diff options
124 files changed, 5505 insertions, 1904 deletions
diff --git a/Documentation/controllers/cpuacct.txt b/Documentation/controllers/cpuacct.txt new file mode 100644 index 00000000000..bb775fbe43d --- /dev/null +++ b/Documentation/controllers/cpuacct.txt @@ -0,0 +1,32 @@ +CPU Accounting Controller +------------------------- + +The CPU accounting controller is used to group tasks using cgroups and +account the CPU usage of these groups of tasks. + +The CPU accounting controller supports multi-hierarchy groups. An accounting +group accumulates the CPU usage of all of its child groups and the tasks +directly present in its group. + +Accounting groups can be created by first mounting the cgroup filesystem. + +# mkdir /cgroups +# mount -t cgroup -ocpuacct none /cgroups + +With the above step, the initial or the parent accounting group +becomes visible at /cgroups. At bootup, this group includes all the +tasks in the system. /cgroups/tasks lists the tasks in this cgroup. +/cgroups/cpuacct.usage gives the CPU time (in nanoseconds) obtained by +this group which is essentially the CPU time obtained by all the tasks +in the system. + +New accounting groups can be created under the parent group /cgroups. + +# cd /cgroups +# mkdir g1 +# echo $$ > g1 + +The above steps create a new group g1 and move the current shell +process (bash) into it. CPU time consumed by this bash and its children +can be obtained from g1/cpuacct.usage and the same is accumulated in +/cgroups/cpuacct.usage also. diff --git a/Documentation/ftrace.txt b/Documentation/ftrace.txt index 9cc4d685dde..35a78bc6651 100644 --- a/Documentation/ftrace.txt +++ b/Documentation/ftrace.txt @@ -82,7 +82,7 @@ of ftrace. Here is a list of some of the key files: tracer is not adding more data, they will display the same information every time they are read. - iter_ctrl: This file lets the user control the amount of data + trace_options: This file lets the user control the amount of data that is displayed in one of the above output files. @@ -94,10 +94,10 @@ of ftrace. Here is a list of some of the key files: only be recorded if the latency is greater than the value in this file. (in microseconds) - trace_entries: This sets or displays the number of bytes each CPU + buffer_size_kb: This sets or displays the number of kilobytes each CPU buffer can hold. The tracer buffers are the same size for each CPU. The displayed number is the size of the - CPU buffer and not total size of all buffers. The + CPU buffer and not total size of all buffers. The trace buffers are allocated in pages (blocks of memory that the kernel uses for allocation, usually 4 KB in size). If the last page allocated has room for more bytes @@ -316,23 +316,23 @@ The above is mostly meaningful for kernel developers. The rest is the same as the 'trace' file. -iter_ctrl ---------- +trace_options +------------- -The iter_ctrl file is used to control what gets printed in the trace +The trace_options file is used to control what gets printed in the trace output. To see what is available, simply cat the file: - cat /debug/tracing/iter_ctrl + cat /debug/tracing/trace_options print-parent nosym-offset nosym-addr noverbose noraw nohex nobin \ - noblock nostacktrace nosched-tree + noblock nostacktrace nosched-tree nouserstacktrace nosym-userobj To disable one of the options, echo in the option prepended with "no". - echo noprint-parent > /debug/tracing/iter_ctrl + echo noprint-parent > /debug/tracing/trace_options To enable an option, leave off the "no". - echo sym-offset > /debug/tracing/iter_ctrl + echo sym-offset > /debug/tracing/trace_options Here are the available options: @@ -378,6 +378,20 @@ Here are the available options: When a trace is recorded, so is the stack of functions. This allows for back traces of trace sites. + userstacktrace - This option changes the trace. + It records a stacktrace of the current userspace thread. + + sym-userobj - when user stacktrace are enabled, look up which object the + address belongs to, and print a relative address + This is especially useful when ASLR is on, otherwise you don't + get a chance to resolve the address to object/file/line after the app is no + longer running + + The lookup is performed when you read trace,trace_pipe,latency_trace. Example: + + a.out-1623 [000] 40874.465068: /root/a.out[+0x480] <-/root/a.out[+0 +x494] <- /root/a.out[+0x4a8] <- /lib/libc-2.7.so[+0x1e1a6] + sched-tree - TBD (any users??) @@ -1299,41 +1313,29 @@ trace entries ------------- Having too much or not enough data can be troublesome in diagnosing -an issue in the kernel. The file trace_entries is used to modify +an issue in the kernel. The file buffer_size_kb is used to modify the size of the internal trace buffers. The number listed is the number of entries that can be recorded per CPU. To know the full size, multiply the number of possible CPUS with the number of entries. - # cat /debug/tracing/trace_entries -65620 + # cat /debug/tracing/buffer_size_kb +1408 (units kilobytes) Note, to modify this, you must have tracing completely disabled. To do that, echo "nop" into the current_tracer. If the current_tracer is not set to "nop", an EINVAL error will be returned. # echo nop > /debug/tracing/current_tracer - # echo 100000 > /debug/tracing/trace_entries - # cat /debug/tracing/trace_entries -100045 - - -Notice that we echoed in 100,000 but the size is 100,045. The entries -are held in individual pages. It allocates the number of pages it takes -to fulfill the request. If more entries may fit on the last page -then they will be added. - - # echo 1 > /debug/tracing/trace_entries - # cat /debug/tracing/trace_entries -85 - -This shows us that 85 entries can fit in a single page. + # echo 10000 > /debug/tracing/buffer_size_kb + # cat /debug/tracing/buffer_size_kb +10000 (units kilobytes) The number of pages which will be allocated is limited to a percentage of available memory. Allocating too much will produce an error. - # echo 1000000000000 > /debug/tracing/trace_entries + # echo 1000000000000 > /debug/tracing/buffer_size_kb -bash: echo: write error: Cannot allocate memory - # cat /debug/tracing/trace_entries + # cat /debug/tracing/buffer_size_kb 85 diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index e0f346d201e..2919a2e9193 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -750,6 +750,14 @@ and is between 256 and 4096 characters. It is defined in the file parameter will force ia64_sal_cache_flush to call ia64_pal_cache_flush instead of SAL_CACHE_FLUSH. + ftrace=[tracer] + [ftrace] will set and start the specified tracer + as early as possible in order to facilitate early + boot debugging. + + ftrace_dump_on_oops + [ftrace] will dump the trace buffers on oops. + gamecon.map[2|3]= [HW,JOY] Multisystem joystick and NES/SNES/PSX pad support via parallel port (up to 5 devices per port) diff --git a/Documentation/lockstat.txt b/Documentation/lockstat.txt index 4ba4664ce5c..9cb9138f7a7 100644 --- a/Documentation/lockstat.txt +++ b/Documentation/lockstat.txt @@ -71,35 +71,50 @@ Look at the current lock statistics: # less /proc/lock_stat -01 lock_stat version 0.2 +01 lock_stat version 0.3 02 ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 03 class name con-bounces contentions waittime-min waittime-max waittime-total acq-bounces acquisitions holdtime-min holdtime-max holdtime-total 04 ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- 05 -06 &inode->i_data.tree_lock-W: 15 21657 0.18 1093295.30 11547131054.85 58 10415 0.16 87.51 6387.60 -07 &inode->i_data.tree_lock-R: 0 0 0.00 0.00 0.00 23302 231198 0.25 8.45 98023.38 -08 -------------------------- -09 &inode->i_data.tree_lock 0 [<ffffffff8027c08f>] add_to_page_cache+0x5f/0x190 -10 -11 ............................................................................................................................................................................................... -12 -13 dcache_lock: 1037 1161 0.38 45.32 774.51 6611 243371 0.15 306.48 77387.24 -14 ----------- -15 dcache_lock 180 [<ffffffff802c0d7e>] sys_getcwd+0x11e/0x230 -16 dcache_lock 165 [<ffffffff802c002a>] d_alloc+0x15a/0x210 -17 dcache_lock 33 [<ffffffff8035818d>] _atomic_dec_and_lock+0x4d/0x70 -18 dcache_lock 1 [<ffffffff802beef8>] shrink_dcache_parent+0x18/0x130 +06 &mm->mmap_sem-W: 233 538 18446744073708 22924.27 607243.51 1342 45806 1.71 8595.89 1180582.34 +07 &mm->mmap_sem-R: 205 587 18446744073708 28403.36 731975.00 1940 412426 0.58 187825.45 6307502.88 +08 --------------- +09 &mm->mmap_sem 487 [<ffffffff8053491f>] do_page_fault+0x466/0x928 +10 &mm->mmap_sem 179 [<ffffffff802a6200>] sys_mprotect+0xcd/0x21d +11 &mm->mmap_sem 279 [<ffffffff80210a57>] sys_mmap+0x75/0xce +12 &mm->mmap_sem 76 [<ffffffff802a490b>] sys_munmap+0x32/0x59 +13 --------------- +14 &mm->mmap_sem 270 [<ffffffff80210a57>] sys_mmap+0x75/0xce +15 &mm->mmap_sem 431 [<ffffffff8053491f>] do_page_fault+0x466/0x928 +16 &mm->mmap_sem 138 [<ffffffff802a490b>] sys_munmap+0x32/0x59 +17 &mm->mmap_sem 145 [<ffffffff802a6200>] sys_mprotect+0xcd/0x21d +18 +19 ............................................................................................................................................................................................... +20 +21 dcache_lock: 621 623 0.52 118.26 1053.02 6745 91930 0.29 316.29 118423.41 +22 ----------- +23 dcache_lock 179 [<ffffffff80378274>] _atomic_dec_and_lock+0x34/0x54 +24 dcache_lock 113 [<ffffffff802cc17b>] d_alloc+0x19a/0x1eb +25 dcache_lock 99 [<ffffffff802ca0dc>] d_rehash+0x1b/0x44 +26 dcache_lock 104 [<ffffffff802cbca0>] d_instantiate+0x36/0x8a +27 ----------- +28 dcache_lock 192 [<ffffffff80378274>] _atomic_dec_and_lock+0x34/0x54 +29 dcache_lock 98 [<ffffffff802ca0dc>] d_rehash+0x1b/0x44 +30 dcache_lock 72 [<ffffffff802cc17b>] d_alloc+0x19a/0x1eb +31 dcache_lock 112 [<ffffffff802cbca0>] d_instantiate+0x36/0x8a This excerpt shows the first two lock class statistics. Line 01 shows the output version - each time the format changes this will be updated. Line 02-04 -show the header with column descriptions. Lines 05-10 and 13-18 show the actual +show the header with column descriptions. Lines 05-18 and 20-31 show the actual statistics. These statistics come in two parts; the actual stats separated by a -short separator (line 08, 14) from the contention points. +short separator (line 08, 13) from the contention points. -The first lock (05-10) is a read/write lock, and shows two lines above the +The first lock (05-18) is a read/write lock, and shows two lines above the short separator. The contention points don't match the column descriptors, -they have two: contentions and [<IP>] symbol. +they have two: contentions and [<IP>] symbol. The second set of contention +points are the points we're contending with. +The integer part of the time values is in us. View the top contending locks: diff --git a/Documentation/markers.txt b/Documentation/markers.txt index 089f6138fcd..6d275e4ef38 100644 --- a/Documentation/markers.txt +++ b/Documentation/markers.txt @@ -70,6 +70,20 @@ a printk warning which identifies the inconsistency: "Format mismatch for probe probe_name (format), marker (format)" +Another way to use markers is to simply define the marker without generating any +function call to actually call into the marker. This is useful in combination +with tracepoint probes in a scheme like this : + +void probe_tracepoint_name(unsigned int arg1, struct task_struct *tsk); + +DEFINE_MARKER_TP(marker_eventname, tracepoint_name, probe_tracepoint_name, + "arg1 %u pid %d"); + +notrace void probe_tracepoint_name(unsigned int arg1, struct task_struct *tsk) +{ + struct marker *marker = &GET_MARKER(kernel_irq_entry); + /* write data to trace buffers ... */ +} * Probe / marker example diff --git a/Documentation/scheduler/sched-arch.txt b/Documentation/scheduler/sched-arch.txt index 941615a9769..d43dbcbd163 100644 --- a/Documentation/scheduler/sched-arch.txt +++ b/Documentation/scheduler/sched-arch.txt @@ -8,7 +8,7 @@ Context switch By default, the switch_to arch function is called with the runqueue locked. This is usually not a problem unless switch_to may need to take the runqueue lock. This is usually due to a wake up operation in -the context switch. See include/asm-ia64/system.h for an example. +the context switch. See arch/ia64/include/asm/system.h for an example. To request the scheduler call switch_to with the runqueue unlocked, you must `#define __ARCH_WANT_UNLOCKED_CTXSW` in a header file @@ -23,7 +23,7 @@ disabled. Interrupts may be enabled over the call if it is likely to introduce a significant interrupt latency by adding the line `#define __ARCH_WANT_INTERRUPTS_ON_CTXSW` in the same place as for unlocked context switches. This define also implies -`__ARCH_WANT_UNLOCKED_CTXSW`. See include/asm-arm/system.h for an +`__ARCH_WANT_UNLOCKED_CTXSW`. See arch/arm/include/asm/system.h for an example. diff --git a/Documentation/tracepoints.txt b/Documentation/tracepoints.txt index 5d354e16749..2d42241a25c 100644 --- a/Documentation/tracepoints.txt +++ b/Documentation/tracepoints.txt @@ -3,28 +3,30 @@ Mathieu Desnoyers -This document introduces Linux Kernel Tracepoints and their use. It provides -examples of how to insert tracepoints in the kernel and connect probe functions -to them and provides some examples of probe functions. +This document introduces Linux Kernel Tracepoints and their use. It +provides examples of how to insert tracepoints in the kernel and +connect probe functions to them and provides some examples of probe +functions. * Purpose of tracepoints -A tracepoint placed in code provides a hook to call a function (probe) that you -can provide at runtime. A tracepoint can be "on" (a probe is connected to it) or -"off" (no probe is attached). When a tracepoint is "off" it has no effect, -except for adding a tiny time penalty (checking a condition for a branch) and -space penalty (adding a few bytes for the function call at the end of the -instrumented function and adds a data structure in a separate section). When a -tracepoint is "on", the function you provide is called each time the tracepoint -is executed, in the execution context of the caller. When the function provided -ends its execution, it returns to the caller (continuing from the tracepoint -site). +A tracepoint placed in code provides a hook to call a function (probe) +that you can provide at runtime. A tracepoint can be "on" (a probe is +connected to it) or "off" (no probe is attached). When a tracepoint is +"off" it has no effect, except for adding a tiny time penalty +(checking a condition for a branch) and space penalty (adding a few +bytes for the function call at the end of the instrumented function +and adds a data structure in a separate section). When a tracepoint +is "on", the function you provide is called each time the tracepoint +is executed, in the execution context of the caller. When the function +provided ends its execution, it returns to the caller (continuing from +the tracepoint site). You can put tracepoints at important locations in the code. They are lightweight hooks that can pass an arbitrary number of parameters, -which prototypes are described in a tracepoint declaration placed in a header -file. +which prototypes are described in a tracepoint declaration placed in a +header file. They can be used for tracing and performance accounting. @@ -42,7 +44,7 @@ In include/trace/subsys.h : #include <linux/tracepoint.h> -DEFINE_TRACE(subsys_eventname, +DECLARE_TRACE(subsys_eventname, TPPTOTO(int firstarg, struct task_struct *p), TPARGS(firstarg, p)); @@ -50,6 +52,8 @@ In subsys/file.c (where the tracing statement must be added) : #include <trace/subsys.h> +DEFINE_TRACE(subsys_eventname); + void somefct(void) { ... @@ -61,31 +65,41 @@ Where : - subsys_eventname is an identifier unique to your event - subsys is the name of your subsystem. - eventname is the name of the event to trace. -- TPPTOTO(int firstarg, struct task_struct *p) is the prototype of the function - called by this tracepoint. -- TPARGS(firstarg, p) are the parameters names, same as found in the prototype. -Connecting a function (probe) to a tracepoint is done by providing a probe -(function to call) for the specific tracepoint through -register_trace_subsys_eventname(). Removing a probe is done through -unregister_trace_subsys_eventname(); it will remove the probe sure there is no -caller left using the probe when it returns. Probe removal is preempt-safe -because preemption is disabled around the probe call. See the "Probe example" -section below for a sample probe module. - -The tracepoint mechanism supports inserting multiple instances of the same -tracepoint, but a single definition must be made of a given tracepoint name over -all the kernel to make sure no type conflict will occur. Name mangling of the -tracepoints is done using the prototypes to make sure typing is correct. -Verification of probe type correctness is done at the registration site by the -compiler. Tracepoints can be put in inline functions, inlined static functions, -and unrolled loops as well as regular functions. - -The naming scheme "subsys_event" is suggested here as a convention intended -to limit collisions. Tracepoint names are global to the kernel: they are -considered as being the same whether they are in the core kernel image or in -modules. +- TPPTOTO(int firstarg, struct task_struct *p) is the prototype of the + function called by this tracepoint. +- TPARGS(firstarg, p) are the parameters names, same as found in the + prototype. + +Connecting a function (probe) to a tracepoint is done by providing a +probe (function to call) for the specific tracepoint through +register_trace_subsys_eventname(). Removing a probe is done through +unregister_trace_subsys_eventname(); it will remove the probe. + +tracepoint_synchronize_unregister() must be called before the end of +the module exit function to make sure there is no caller left using +the probe. This, and the fact that preemption is disabled around the +probe call, make sure that probe removal and module unload are safe. +See the "Probe example" section below for a sample probe module. + +The tracepoint mechanism supports inserting multiple instances of the +same tracepoint, but a single definition must be made of a given +tracepoint name over all the kernel to make sure no type conflict will +occur. Name mangling of the tracepoints is done using the prototypes +to make sure typing is correct. Verification of probe type correctness +is done at the registration site by the compiler. Tracepoints can be +put in inline functions, inlined static functions, and unrolled loops +as well as regular functions. + +The naming scheme "subsys_event" is suggested here as a convention +intended to limit collisions. Tracepoint names are global to the +kernel: they are considered as being the same whether they are in the +core kernel image or in modules. + +If the tracepoint has to be used in kernel modules, an +EXPORT_TRACEPOINT_SYMBOL_GPL() or EXPORT_TRACEPOINT_SYMBOL() can be +used to export the defined tracepoints. * Probe / tracepoint example diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 6bd91ed7cd0..7fa8f615ba6 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -99,7 +99,7 @@ config GENERIC_IOMAP bool default y -config SCHED_NO_NO_OMIT_FRAME_POINTER +config SCHED_OMIT_FRAME_POINTER bool default y diff --git a/arch/ia64/include/asm/topology.h b/arch/ia64/include/asm/topology.h index 35bcb641c9e..a3cc9f65f95 100644 --- a/arch/ia64/include/asm/topology.h +++ b/arch/ia64/include/asm/topology.h @@ -55,7 +55,6 @@ void build_cpu_to_node_map(void); #define SD_CPU_INIT (struct sched_domain) { \ - .span = CPU_MASK_NONE, \ .parent = NULL, \ .child = NULL, \ .groups = NULL, \ @@ -80,7 +79,6 @@ void build_cpu_to_node_map(void); /* sched_domains SD_NODE_INIT for IA64 NUMA machines */ #define SD_NODE_INIT (struct sched_domain) { \ - .span = CPU_MASK_NONE, \ .parent = NULL, \ .child = NULL, \ .groups = NULL, \ diff --git a/arch/m32r/Kconfig b/arch/m32r/Kconfig index dbaed4a6381..29047d5c259 100644 --- a/arch/m32r/Kconfig +++ b/arch/m32r/Kconfig @@ -273,7 +273,7 @@ config GENERIC_CALIBRATE_DELAY bool default y -config SCHED_NO_NO_OMIT_FRAME_POINTER +config SCHED_OMIT_FRAME_POINTER bool default y diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig index f4af967a6b3..a5255e7c79e 100644 --- a/arch/mips/Kconfig +++ b/arch/mips/Kconfig @@ -653,7 +653,7 @@ config GENERIC_CMOS_UPDATE bool default y -config SCHED_NO_NO_OMIT_FRAME_POINTER +config SCHED_OMIT_FRAME_POINTER bool default y diff --git a/arch/mips/include/asm/mach-ip27/topology.h b/arch/mips/include/asm/mach-ip27/topology.h index 7785bec732f..1fb959f9898 100644 --- a/arch/mips/include/asm/mach-ip27/topology.h +++ b/arch/mips/include/asm/mach-ip27/topology.h @@ -37,7 +37,6 @@ extern unsigned char __node_distances[MAX_COMPACT_NODES][MAX_COMPACT_NODES]; /* sched_domains SD_NODE_INIT for SGI IP27 machines */ #define SD_NODE_INIT (struct sched_domain) { \ - .span = CPU_MASK_NONE, \ .parent = NULL, \ .child = NULL, \ .groups = NULL, \ diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 525c13a4de9..adb23ea1c1e 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -141,7 +141,7 @@ config GENERIC_NVRAM bool default y if PPC32 -config SCHED_NO_NO_OMIT_FRAME_POINTER +config SCHED_OMIT_FRAME_POINTER bool default y diff --git a/arch/powerpc/include/asm/ftrace.h b/arch/powerpc/include/asm/ftrace.h index b298f7a631e..e5f2ae8362f 100644 --- a/arch/powerpc/include/asm/ftrace.h +++ b/arch/powerpc/include/asm/ftrace.h @@ -7,7 +7,19 @@ #ifndef __ASSEMBLY__ extern void _mcount(void); -#endif + +#ifdef CONFIG_DYNAMIC_FTRACE +static inline unsigned long ftrace_call_adjust(unsigned long addr) +{ + /* reloction of mcount call site is the same as the address */ + return addr; +} + +struct dyn_arch_ftrace { + struct module *mod; +}; +#endif /* CONFIG_DYNAMIC_FTRACE */ +#endif /* __ASSEMBLY__ */ #endif diff --git a/arch/powerpc/include/asm/module.h b/arch/powerpc/include/asm/module.h index e5f14b13ccf..08454880a2c 100644 --- a/arch/powerpc/include/asm/module.h +++ b/arch/powerpc/include/asm/module.h @@ -34,11 +34,19 @@ struct mod_arch_specific { #ifdef __powerpc64__ unsigned int stubs_section; /* Index of stubs section in module */ unsigned int toc_section; /* What section is the TOC? */ -#else +#ifdef CONFIG_DYNAMIC_FTRACE + unsigned long toc; + unsigned long tramp; +#endif + +#else /* powerpc64 */ /* Indices of PLT sections within module. */ unsigned int core_plt_section; unsigned int init_plt_section; +#ifdef CONFIG_DYNAMIC_FTRACE + unsigned long tramp; #endif +#endif /* powerpc64 */ /* List of BUG addresses, source line numbers and filenames */ struct list_head bug_list; @@ -68,6 +76,12 @@ struct mod_arch_specific { # endif /* MODULE */ #endif +#ifdef CONFIG_DYNAMIC_FTRACE +# ifdef MODULE + asm(".section .ftrace.tramp,\"ax\",@nobits; .align 3; .previous"); +# endif /* MODULE */ +#endif + struct exception_table_entry; void sort_ex_table(struct exception_table_entry *start, diff --git a/arch/powerpc/include/asm/topology.h b/arch/powerpc/include/asm/topology.h index c32da6f9799..373fca394a5 100644 --- a/arch/powerpc/include/asm/topology.h +++ b/arch/powerpc/include/asm/topology.h @@ -48,7 +48,6 @@ static inline int pcibus_to_node(struct pci_bus *bus) /* sched_domains SD_NODE_INIT for PPC64 machines */ #define SD_NODE_INIT (struct sched_domain) { \ - .span = CPU_MASK_NONE, \ .parent = NULL, \ .child = NULL, \ .groups = NULL, \ diff --git a/arch/powerpc/kernel/ftrace.c b/arch/powerpc/kernel/ftrace.c index f4b006ed0ab..3271cd698e4 100644 --- a/arch/powerpc/kernel/ftrace.c +++ b/arch/powerpc/kernel/ftrace.c @@ -9,22 +9,30 @@ #include <linux/spinlock.h> #include <linux/hardirq.h> +#include <linux/uaccess.h> +#include <linux/module.h> #include <linux/ftrace.h> #include <linux/percpu.h> #include <linux/init.h> #include <linux/list.h> #include <asm/cacheflush.h> +#include <asm/code-patching.h> #include <asm/ftrace.h> +#if 0 +#define DEBUGP printk +#else +#define DEBUGP(fmt , ...) do { } while (0) +#endif -static unsigned int ftrace_nop = 0x60000000; +static unsigned int ftrace_nop = PPC_NOP_INSTR; #ifdef CONFIG_PPC32 # define GET_ADDR(addr) addr #else /* PowerPC64's functions are data that points to the functions */ -# define GET_ADDR(addr) *(unsigned long *)addr +# define GET_ADDR(addr) (*(unsigned long *)addr) #endif @@ -33,12 +41,12 @@ static unsigned int ftrace_calc_offset(long ip, long addr) return (int)(addr - ip); } -unsigned char *ftrace_nop_replace(void) +static unsigned char *ftrace_nop_replace(void) { return (char *)&ftrace_nop; } -unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr) +static unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr) { static unsigned int op; @@ -68,49 +76,434 @@ unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr) # define _ASM_PTR " .long " #endif -int +static int ftrace_modify_code(unsigned long ip, unsigned char *old_code, unsigned char *new_code) { - unsigned replaced; - unsigned old = *(unsigned *)old_code; - unsigned new = *(unsigned *)new_code; - int faulted = 0; + unsigned char replaced[MCOUNT_INSN_SIZE]; /* * Note: Due to modules and __init, code can * disappear and change, we need to protect against faulting - * as well as code changing. + * as well as code changing. We do this by using the + * probe_kernel_* functions. * * No real locking needed, this code is run through - * kstop_machine. + * kstop_machine, or before SMP starts. */ - asm volatile ( - "1: lwz %1, 0(%2)\n" - " cmpw %1, %5\n" - " bne 2f\n" - " stwu %3, 0(%2)\n" - "2:\n" - ".section .fixup, \"ax\"\n" - "3: li %0, 1\n" - " b 2b\n" - ".previous\n" - ".section __ex_table,\"a\"\n" - _ASM_ALIGN "\n" - _ASM_PTR "1b, 3b\n" - ".previous" - : "=r"(faulted), "=r"(replaced) - : "r"(ip), "r"(new), - "0"(faulted), "r"(old) - : "memory"); - - if (replaced != old && replaced != new) - faulted = 2; - - if (!faulted) - flush_icache_range(ip, ip + 8); - - return faulted; + + /* read the text we want to modify */ + if (probe_kernel_read(replaced, (void *)ip, MCOUNT_INSN_SIZE)) + return -EFAULT; + + /* Make sure it is what we expect it to be */ + if (memcmp(replaced, old_code, MCOUNT_INSN_SIZE) != 0) + return -EINVAL; + + /* replace the text with the new text */ + if (probe_kernel_write((void *)ip, new_code, MCOUNT_INSN_SIZE)) + return -EPERM; + + flush_icache_range(ip, ip + 8); + + return 0; +} + +/* + * Helper functions that are the same for both PPC64 and PPC32. + */ +static int test_24bit_addr(unsigned long ip, unsigned long addr) +{ + long diff; + + /* + * Can we get to addr from ip in 24 bits? + * (26 really, since we mulitply by 4 for 4 byte alignment) + */ + diff = addr - ip; + + /* + * Return true if diff is less than 1 << 25 + * and greater than -1 << 26. + */ + return (diff < (1 << 25)) && (diff > (-1 << 26)); +} + +static int is_bl_op(unsigned int op) +{ + return (op & 0xfc000003) == 0x48000001; +} + +static int test_offset(unsigned long offset) +{ + return (offset + 0x2000000 > 0x3ffffff) || ((offset & 3) != 0); +} + +static unsigned long find_bl_target(unsigned long ip, unsigned int op) +{ + static int offset; + + offset = (op & 0x03fffffc); + /* make it signed */ + if (offset & 0x02000000) + offset |= 0xfe000000; + + return ip + (long)offset; +} + +static unsigned int branch_offset(unsigned long offset) +{ + /* return "bl ip+offset" */ + return 0x48000001 | (offset & 0x03fffffc); +} + +#ifdef CONFIG_PPC64 +static int +__ftrace_make_nop(struct module *mod, + struct dyn_ftrace *rec, unsigned long addr) +{ + unsigned char replaced[MCOUNT_INSN_SIZE * 2]; + unsigned int *op = (unsigned *)&replaced; + unsigned char jmp[8]; + unsigned long *ptr = (unsigned long *)&jmp; + unsigned long ip = rec->ip; + unsigned long tramp; + int offset; + + /* read where this goes */ + if (probe_kernel_read(replaced, (void *)ip, MCOUNT_INSN_SIZE)) + return -EFAULT; + + /* Make sure that that this is still a 24bit jump */ + if (!is_bl_op(*op)) { + printk(KERN_ERR "Not expected bl: opcode is %x\n", *op); + return -EINVAL; + } + + /* lets find where the pointer goes */ + tramp = find_bl_target(ip, *op); + + /* + * On PPC64 the trampoline looks like: + * 0x3d, 0x82, 0x00, 0x00, addis r12,r2, <high> + * 0x39, 0x8c, 0x00, 0x00, addi r12,r12, <low> + * Where the bytes 2,3,6 and 7 make up the 32bit offset + * to the TOC that holds the pointer. + * to jump to. + * 0xf8, 0x41, 0x00, 0x28, std r2,40(r1) + * 0xe9, 0x6c, 0x00, 0x20, ld r11,32(r12) + * The actually address is 32 bytes from the offset + * into the TOC. + * 0xe8, 0x4c, 0x00, 0x28, ld r2,40(r12) + */ + + DEBUGP("ip:%lx jumps to %lx r2: %lx", ip, tramp, mod->arch.toc); + + /* Find where the trampoline jumps to */ + if (probe_kernel_read(jmp, (void *)tramp, 8)) { + printk(KERN_ERR "Failed to read %lx\n", tramp); + return -EFAULT; + } + + DEBUGP(" %08x %08x", + (unsigned)(*ptr >> 32), + (unsigned)*ptr); + + offset = (unsigned)jmp[2] << 24 | + (unsigned)jmp[3] << 16 | + (unsigned)jmp[6] << 8 | + (unsigned)jmp[7]; + + DEBUGP(" %x ", offset); + + /* get the address this jumps too */ + tramp = mod->arch.toc + offset + 32; + DEBUGP("toc: %lx", tramp); + + if (probe_kernel_read(jmp, (void *)tramp, 8)) { + printk(KERN_ERR "Failed to read %lx\n", tramp); + return -EFAULT; + } + + DEBUGP(" %08x %08x\n", + (unsigned)(*ptr >> 32), + (unsigned)*ptr); + + /* This should match what was called */ + if (*ptr != GET_ADDR(addr)) { + printk(KERN_ERR "addr does not match %lx\n", *ptr); + return -EINVAL; + } + + /* + * We want to nop the line, but the next line is + * 0xe8, 0x41, 0x00, 0x28 ld r2,40(r1) + * This needs to be turned to a nop too. + */ + if (probe_kernel_read(replaced, (void *)(ip+4), MCOUNT_INSN_SIZE)) + return -EFAULT; + + if (*op != 0xe8410028) { + printk(KERN_ERR "Next line is not ld! (%08x)\n", *op); + return -EINVAL; + } + + /* + * Milton Miller pointed out that we can not blindly do nops. + * If a task was preempted when calling a trace function, + * the nops will remove the way to restore the TOC in r2 + * and the r2 TOC will get corrupted. + */ + + /* + * Replace: + * bl <tramp> <==== will be replaced with "b 1f" + * ld r2,40(r1) + * 1: + */ + op[0] = 0x48000008; /* b +8 */ + + if (probe_kernel_write((void *)ip, replaced, MCOUNT_INSN_SIZE)) + return -EPERM; + + return 0; +} + +#else /* !PPC64 */ +static int +__ftrace_make_nop(struct module *mod, + struct dyn_ftrace *rec, unsigned long addr) +{ + unsigned char replaced[MCOUNT_INSN_SIZE]; + unsigned int *op = (unsigned *)&replaced; + unsigned char jmp[8]; + unsigned int *ptr = (unsigned int *)&jmp; + unsigned long ip = rec->ip; + unsigned long tramp; + int offset; + + if (probe_kernel_read(replaced, (void *)ip, MCOUNT_INSN_SIZE)) + return -EFAULT; + + /* Make sure that that this is still a 24bit jump */ + if (!is_bl_op(*op)) { + printk(KERN_ERR "Not expected bl: opcode is %x\n", *op); + return -EINVAL; + } + + /* lets find where the pointer goes */ + tramp = find_bl_target(ip, *op); + + /* + * On PPC32 the trampoline looks like: + * lis r11,sym@ha + * addi r11,r11,sym@l + * mtctr r11 + * bctr + */ + + DEBUGP("ip:%lx jumps to %lx", ip, tramp); + + /* Find where the trampoline jumps to */ + if (probe_kernel_read(jmp, (void *)tramp, 8)) { + printk(KERN_ERR "Failed to read %lx\n", tramp); + return -EFAULT; + } + + DEBUGP(" %08x %08x ", ptr[0], ptr[1]); + + tramp = (ptr[1] & 0xffff) | + ((ptr[0] & 0xffff) << 16); + if (tramp & 0x8000) + tramp -= 0x10000; + + DEBUGP(" %x ", tramp); + + if (tramp != addr) { + printk(KERN_ERR + "Trampoline location %08lx does not match addr\n", + tramp); + return -EINVAL; + } + + op[0] = PPC_NOP_INSTR; + + if (probe_kernel_write((void *)ip, replaced, MCOUNT_INSN_SIZE)) + return -EPERM; + + return 0; +} +#endif /* PPC64 */ + +int ftrace_make_nop(struct module *mod, + struct dyn_ftrace *rec, unsigned long addr) +{ + unsigned char *old, *new; + unsigned long ip = rec->ip; + + /* + * If the calling address is more that 24 bits away, + * then we had to use a trampoline to make the call. + * Otherwise just update the call site. + */ + if (test_24bit_addr(ip, addr)) { + /* within range */ + old = ftrace_call_replace(ip, addr); + new = ftrace_nop_replace(); + return ftrace_modify_code(ip, old, new); + } + + /* + * Out of range jumps are called from modules. + * We should either already have a pointer to the module + * or it has been passed in. + */ + if (!rec->arch.mod) { + if (!mod) { + printk(KERN_ERR "No module loaded addr=%lx\n", + addr); + return -EFAULT; + } + rec->arch.mod = mod; + } else if (mod) { + if (mod != rec->arch.mod) { + printk(KERN_ERR + "Record mod %p not equal to passed in mod %p\n", + rec->arch.mod, mod); + return -EINVAL; + } + /* nothing to do if mod == rec->arch.mod */ + } else + mod = rec->arch.mod; + + return __ftrace_make_nop(mod, rec, addr); + +} + +#ifdef CONFIG_PPC64 +static int +__ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) +{ + unsigned char replaced[MCOUNT_INSN_SIZE * 2]; + unsigned int *op = (unsigned *)&replaced; + unsigned long ip = rec->ip; + unsigned long offset; + + /* read where this goes */ + if (probe_kernel_read(replaced, (void *)ip, MCOUNT_INSN_SIZE * 2)) + return -EFAULT; + + /* + * It should be pointing to two nops or + * b +8; ld r2,40(r1) + */ + if (((op[0] != 0x48000008) || (op[1] != 0xe8410028)) && + ((op[0] != PPC_NOP_INSTR) || (op[1] != PPC_NOP_INSTR))) { + printk(KERN_ERR "Expected NOPs but have %x %x\n", op[0], op[1]); + return -EINVAL; + } + + /* If we never set up a trampoline to ftrace_caller, then bail */ + if (!rec->arch.mod->arch.tramp) { + printk(KERN_ERR "No ftrace trampoline\n"); + return -EINVAL; + } + + /* now calculate a jump to the ftrace caller trampoline */ + offset = rec->arch.mod->arch.tramp - ip; + + if (test_offset(offset)) { + printk(KERN_ERR "REL24 %li out of range!\n", + (long int)offset); + return -EINVAL; + } + + /* Set to "bl addr" */ + op[0] = branch_offset(offset); + /* ld r2,40(r1) */ + op[1] = 0xe8410028; + + DEBUGP("write to %lx\n", rec->ip); + + if (probe_kernel_write((void *)ip, replaced, MCOUNT_INSN_SIZE * 2)) + return -EPERM; + + return 0; +} +#else +static int +__ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) +{ + unsigned char replaced[MCOUNT_INSN_SIZE]; + unsigned int *op = (unsigned *)&replaced; + unsigned long ip = rec->ip; + unsigned long offset; + + /* read where this goes */ + if (probe_kernel_read(replaced, (void *)ip, MCOUNT_INSN_SIZE)) + return -EFAULT; + + /* It should be pointing to a nop */ + if (op[0] != PPC_NOP_INSTR) { + printk(KERN_ERR "Expected NOP but have %x\n", op[0]); + return -EINVAL; + } + + /* If we never set up a trampoline to ftrace_caller, then bail */ + if (!rec->arch.mod->arch.tramp) { + printk(KERN_ERR "No ftrace trampoline\n"); + return -EINVAL; + } + + /* now calculate a jump to the ftrace caller trampoline */ + offset = rec->arch.mod->arch.tramp - ip; + + if (test_offset(offset)) { + printk(KERN_ERR "REL24 %li out of range!\n", + (long int)offset); + return -EINVAL; + } + + /* Set to "bl addr" */ + op[0] = branch_offset(offset); + + DEBUGP("write to %lx\n", rec->ip); + + if (probe_kernel_write((void *)ip, replaced, MCOUNT_INSN_SIZE)) + return -EPERM; + + return 0; +} +#endif /* CONFIG_PPC64 */ + +int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) +{ + unsigned char *old, *new; + unsigned long ip = rec->ip; + + /* + * If the calling address is more that 24 bits away, + * then we had to use a trampoline to make the call. + * Otherwise just update the call site. + */ + if (test_24bit_addr(ip, addr)) { + /* within range */ + old = ftrace_nop_replace(); + new = ftrace_call_replace(ip, addr); + return ftrace_modify_code(ip, old, new); + } + + /* + * Out of range jumps are called from modules. + * Being that we are converting from nop, it had better + * already have a module defined. + */ + if (!rec->arch.mod) { + printk(KERN_ERR "No module loaded\n"); + return -EINVAL; + } + + return __ftrace_make_call(rec, addr); } int ftrace_update_ftrace_func(ftrace_func_t func) @@ -128,10 +521,10 @@ int ftrace_update_ftrace_func(ftrace_func_t func) int __init ftrace_dyn_arch_init(void *data) { - /* This is running in kstop_machine */ + /* caller expects data to be zero */ + unsigned long *p = data; - ftrace_mcount_set(data); + *p = 0; return 0; } - diff --git a/arch/powerpc/kernel/idle.c b/arch/powerpc/kernel/idle.c index 31982d05d81..88d9c1d5e5f 100644 --- a/arch/powerpc/kernel/idle.c +++ b/arch/powerpc/kernel/idle.c @@ -69,10 +69,15 @@ void cpu_idle(void) smp_mb(); local_irq_disable(); + /* Don't trace irqs off for idle */ + stop_critical_timings(); + /* check again after disabling irqs */ if (!need_resched() && !cpu_should_die()) ppc_md.power_save(); + start_critical_timings(); + local_irq_enable(); set_thread_flag(TIF_POLLING_NRFLAG); diff --git a/arch/powerpc/kernel/module_32.c b/arch/powerpc/kernel/module_32.c index 2df91a03462..f832773fc28 100644 --- a/arch/powerpc/kernel/module_32.c +++ b/arch/powerpc/kernel/module_32.c @@ -22,6 +22,7 @@ #include <linux/fs.h> #include <linux/string.h> #include <linux/kernel.h> +#include <linux/ftrace.h> #include <linux/cache.h> #include <linux/bug.h> #include <linux/sort.h> @@ -53,6 +54,9 @@ static unsigned int count_relocs(const Elf32_Rela *rela, unsigned int num) r_addend = rela[i].r_addend; } +#ifdef CONFIG_DYNAMIC_FTRACE + _count_relocs++; /* add one for ftrace_caller */ +#endif return _count_relocs; } @@ -306,5 +310,11 @@ int apply_relocate_add(Elf32_Shdr *sechdrs, return -ENOEXEC; } } +#ifdef CONFIG_DYNAMIC_FTRACE + module->arch.tramp = + do_plt_call(module->module_core, + (unsigned long)ftrace_caller, + sechdrs, module); +#endif return 0; } diff --git a/arch/powerpc/kernel/module_64.c b/arch/powerpc/kernel/module_64.c index 1af2377e499..8992b031a7b 100644 --- a/arch/powerpc/kernel/module_64.c +++ b/arch/powerpc/kernel/module_64.c @@ -20,6 +20,7 @@ #include <linux/moduleloader.h> #include <linux/err.h> #include <linux/vmalloc.h> +#include <linux/ftrace.h> #include <linux/bug.h> #include <asm/module.h> #include <asm/firmware.h> @@ -163,6 +164,11 @@ static unsigned long get_stubs_size(const Elf64_Ehdr *hdr, } } +#ifdef CONFIG_DYNAMIC_FTRACE + /* make the trampoline to the ftrace_caller */ + relocs++; +#endif + DEBUGP("Looks like a total of %lu stubs, max\n", relocs); return relocs * sizeof(struct ppc64_stub_entry); } @@ -441,5 +447,12 @@ int apply_relocate_add(Elf64_Shdr *sechdrs, } } +#ifdef CONFIG_DYNAMIC_FTRACE + me->arch.toc = my_r2(sechdrs, me); + me->arch.tramp = stub_for_addr(sechdrs, + (unsigned long)ftrace_caller, + me); +#endif + return 0; } diff --git a/arch/sh/include/asm/topology.h b/arch/sh/include/asm/topology.h index 95f0085e098..279d9cc4a00 100644 --- a/arch/sh/include/asm/topology.h +++ b/arch/sh/include/asm/topology.h @@ -5,7 +5,6 @@ /* sched_domains SD_NODE_INIT for sh machines */ #define SD_NODE_INIT (struct sched_domain) { \ - .span = CPU_MASK_NONE, \ .parent = NULL, \ .child = NULL, \ .groups = NULL, \ diff --git a/arch/um/include/asm/system.h b/arch/um/include/asm/system.h index 753346e2cdf..ae5f94d6317 100644 --- a/arch/um/include/asm/system.h +++ b/arch/um/include/asm/system.h @@ -11,21 +11,21 @@ extern int get_signals(void); extern void block_signals(void); extern void unblock_signals(void); -#define local_save_flags(flags) do { typecheck(unsigned long, flags); \ +#define raw_local_save_flags(flags) do { typecheck(unsigned long, flags); \ (flags) = get_signals(); } while(0) -#define local_irq_restore(flags) do { typecheck(unsigned long, flags); \ +#define raw_local_irq_restore(flags) do { typecheck(unsigned long, flags); \ set_signals(flags); } while(0) -#define local_irq_save(flags) do { local_save_flags(flags); \ - local_irq_disable(); } while(0) +#define raw_local_irq_save(flags) do { raw_local_save_flags(flags); \ + raw_local_irq_disable(); } while(0) -#define local_irq_enable() unblock_signals() -#define local_irq_disable() block_signals() +#define raw_local_irq_enable() unblock_signals() +#define raw_local_irq_disable() block_signals() #define irqs_disabled() \ ({ \ unsigned long flags; \ - local_save_flags(flags); \ + raw_local_save_flags(flags); \ (flags == 0); \ }) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index ac22bb7719f..352f63df1d8 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -29,11 +29,14 @@ config X86 select HAVE_FTRACE_MCOUNT_RECORD select HAVE_DYNAMIC_FTRACE select HAVE_FUNCTION_TRACER + select HAVE_FUNCTION_RET_TRACER if X86_32 + select HAVE_FUNCTION_TRACE_MCOUNT_TEST select HAVE_KVM if ((X86_32 && !X86_VOYAGER && !X86_VISWS && !X86_NUMAQ) || X86_64) select HAVE_ARCH_KGDB if !X86_VOYAGER select HAVE_ARCH_TRACEHOOK select HAVE_GENERIC_DMA_COHERENT if X86_32 select HAVE_EFFICIENT_UNALIGNED_ACCESS + select USER_STACKTRACE_SUPPORT config ARCH_DEFCONFIG string @@ -367,10 +370,10 @@ config X86_RDC321X as R-8610-(G). If you don't have one of these chips, you should say N here. -config SCHED_NO_NO_OMIT_FRAME_POINTER +config SCHED_OMIT_FRAME_POINTER def_bool y prompt "Single-depth WCHAN output" - depends on X86_32 + depends on X86 help Calculate simpler /proc/<PID>/wchan values. If this option is disabled then wchan values will recurse back to the diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index 2a3dfbd5e67..fa013f529b7 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -186,14 +186,10 @@ config IOMMU_LEAK Add a simple leak tracer to the IOMMU code. This is useful when you are debugging a buggy device driver that leaks IOMMU mappings. -config MMIOTRACE_HOOKS - bool - config MMIOTRACE bool "Memory mapped IO tracing" depends on DEBUG_KERNEL && PCI select TRACING - select MMIOTRACE_HOOKS help Mmiotrace traces Memory Mapped I/O access and is meant for debugging and reverse engineering. It is called from the ioremap diff --git a/arch/x86/include/asm/ftrace.h b/arch/x86/include/asm/ftrace.h index 9e8bc29b8b1..754a3e082f9 100644 --- a/arch/x86/include/asm/ftrace.h +++ b/arch/x86/include/asm/ftrace.h @@ -17,8 +17,40 @@ static inline unsigned long ftrace_call_adjust(unsigned long addr) */ return addr - 1; } -#endif +#ifdef CONFIG_DYNAMIC_FTRACE + +struct dyn_arch_ftrace { + /* No extra data needed for x86 */ +}; + +#endif /* CONFIG_DYNAMIC_FTRACE */ +#endif /* __ASSEMBLY__ */ #endif /* CONFIG_FUNCTION_TRACER */ +#ifdef CONFIG_FUNCTION_RET_TRACER + +#ifndef __ASSEMBLY__ + +/* + * Stack of return addresses for functions + * of a thread. + * Used in struct thread_info + */ +struct ftrace_ret_stack { + unsigned long ret; + unsigned long func; + unsigned long long calltime; +}; + +/* + * Primary handler of a function return. + * It relays on ftrace_return_to_handler. + * Defined in entry32.S + */ +extern void return_to_handler(void); + +#endif /* __ASSEMBLY__ */ +#endif /* CONFIG_FUNCTION_RET_TRACER */ + #endif /* _ASM_X86_FTRACE_H */ diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h index e44d379faad..0921b4018c1 100644 --- a/arch/x86/include/asm/thread_info.h +++ b/arch/x86/include/asm/thread_info.h @@ -20,6 +20,8 @@ struct task_struct; struct exec_domain; #include <asm/processor.h> +#include <asm/ftrace.h> +#include <asm/atomic.h> struct thread_info { struct task_struct *task; /* main task structure */ diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h index 35c54921b2e..99192bb55a5 100644 --- a/arch/x86/include/asm/uaccess.h +++ b/arch/x86/include/asm/uaccess.h @@ -157,6 +157,7 @@ extern int __get_user_bad(void); int __ret_gu; \ unsigned long __val_gu; \ __chk_user_ptr(ptr); \ + might_fault(); \ switch (sizeof(*(ptr))) { \ case 1: \ __get_user_x(1, __ret_gu, __val_gu, ptr); \ @@ -241,6 +242,7 @@ extern void __put_user_8(void); int __ret_pu; \ __typeof__(*(ptr)) __pu_val; \ __chk_user_ptr(ptr); \ + might_fault(); \ __pu_val = x; \ switch (sizeof(*(ptr))) { \ case 1: \ diff --git a/arch/x86/include/asm/uaccess_32.h b/arch/x86/include/asm/uaccess_32.h index d095a3aeea1..5e06259e90e 100644 --- a/arch/x86/include/asm/uaccess_32.h +++ b/arch/x86/include/asm/uaccess_32.h @@ -82,8 +82,8 @@ __copy_to_user_inatomic(void __user *to, const void *from, unsigned long n) static __always_inline unsigned long __must_check __copy_to_user(void __user *to, const void *from, unsigned long n) { - might_sleep(); - return __copy_to_user_inatomic(to, from, n); + might_fault(); + return __copy_to_user_inatomic(to, from, n); } static __always_inline unsigned long @@ -137,7 +137,7 @@ __copy_from_user_inatomic(void *to, const void __user *from, unsigned long n) static __always_inline unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n) { - might_sleep(); + might_fault(); if (__builtin_constant_p(n)) { unsigned long ret; @@ -159,7 +159,7 @@ __copy_from_user(void *to, const void __user *from, unsigned long n) static __always_inline unsigned long __copy_from_user_nocache(void *to, const void __user *from, unsigned long n) { - might_sleep(); + might_fault(); if (__builtin_constant_p(n)) { unsigned long ret; diff --git a/arch/x86/include/asm/uaccess_64.h b/arch/x86/include/asm/uaccess_64.h index f8cfd00db45..84210c479fc 100644 --- a/arch/x86/include/asm/uaccess_64.h +++ b/arch/x86/include/asm/uaccess_64.h @@ -29,6 +29,8 @@ static __always_inline __must_check int __copy_from_user(void *dst, const void __user *src, unsigned size) { int ret = 0; + + might_fault(); if (!__builtin_constant_p(size)) return copy_user_generic(dst, (__force void *)src, size); switch (size) { @@ -71,6 +73,8 @@ static __always_inline __must_check int __copy_to_user(void __user *dst, const void *src, unsigned size) { int ret = 0; + + might_fault(); if (!__builtin_constant_p(size)) return copy_user_generic((__force void *)dst, src, size); switch (size) { @@ -113,6 +117,8 @@ static __always_inline __must_check int __copy_in_user(void __user *dst, const void __user *src, unsigned size) { int ret = 0; + + might_fault(); if (!__builtin_constant_p(size)) return copy_user_generic((__force void *)dst, (__force void *)src, size); diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index b62a7667828..af2bc36ca1c 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -14,6 +14,11 @@ CFLAGS_REMOVE_paravirt-spinlocks.o = -pg CFLAGS_REMOVE_ftrace.o = -pg endif +ifdef CONFIG_FUNCTION_RET_TRACER +# Don't trace __switch_to() but let it for function tracer +CFLAGS_REMOVE_process_32.o = -pg +endif + # # vsyscalls (which work on the user stack) should have # no stack-protector checks: @@ -65,6 +70,7 @@ obj-$(CONFIG_X86_LOCAL_APIC) += apic.o nmi.o obj-$(CONFIG_X86_IO_APIC) += io_apic.o obj-$(CONFIG_X86_REBOOTFIXUPS) += reboot_fixups_32.o obj-$(CONFIG_DYNAMIC_FTRACE) += ftrace.o +obj-$(CONFIG_FUNCTION_RET_TRACER) += ftrace.o obj-$(CONFIG_KEXEC) += machine_kexec_$(BITS).o obj-$(CONFIG_KEXEC) += relocate_kernel_$(BITS).o crash.o obj-$(CONFIG_CRASH_DUMP) += crash_dump_$(BITS).o diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index 28b597ef9ca..74defe21ba4 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -1157,6 +1157,9 @@ ENTRY(mcount) END(mcount) ENTRY(ftrace_caller) + cmpl $0, function_trace_stop + jne ftrace_stub + pushl %eax pushl %ecx pushl %edx @@ -1180,8 +1183,15 @@ END(ftrace_caller) #else /* ! CONFIG_DYNAMIC_FTRACE */ ENTRY(mcount) + cmpl $0, function_trace_stop + jne ftrace_stub + cmpl $ftrace_stub, ftrace_trace_function jnz trace +#ifdef CONFIG_FUNCTION_RET_TRACER + cmpl $ftrace_stub, ftrace_function_return + jnz ftrace_return_caller +#endif .globl ftrace_stub ftrace_stub: ret @@ -1200,12 +1210,42 @@ trace: popl %edx popl %ecx popl %eax - jmp ftrace_stub END(mcount) #endif /* CONFIG_DYNAMIC_FTRACE */ #endif /* CONFIG_FUNCTION_TRACER */ +#ifdef CONFIG_FUNCTION_RET_TRACER +ENTRY(ftrace_return_caller) + cmpl $0, function_trace_stop + jne ftrace_stub + + pushl %eax + pushl %ecx + pushl %edx + movl 0xc(%esp), %edx + lea 0x4(%ebp), %eax + call prepare_ftrace_return + popl %edx + popl %ecx + popl %eax + ret +END(ftrace_return_caller) + +.globl return_to_handler +return_to_handler: + pushl $0 + pushl %eax + pushl %ecx + pushl %edx + call ftrace_return_to_handler + movl %eax, 0xc(%esp) + popl %edx + popl %ecx + popl %eax + ret +#endif + .section .rodata,"a" #include "syscall_table_32.S" diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index b86f332c96a..08aa6b10933 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -68,6 +68,8 @@ ENTRY(mcount) END(mcount) ENTRY(ftrace_caller) + cmpl $0, function_trace_stop + jne ftrace_stub /* taken from glibc */ subq $0x38, %rsp @@ -103,6 +105,9 @@ END(ftrace_caller) #else /* ! CONFIG_DYNAMIC_FTRACE */ ENTRY(mcount) + cmpl $0, function_trace_stop + jne ftrace_stub + cmpq $ftrace_stub, ftrace_trace_function jnz trace .globl ftrace_stub diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index 50ea0ac8c9b..bb137f7297e 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -14,14 +14,17 @@ #include <linux/uaccess.h> #include <linux/ftrace.h> #include <linux/percpu.h> +#include <linux/sched.h> #include <linux/init.h> #include <linux/list.h> #include <asm/ftrace.h> +#include <linux/ftrace.h> #include <asm/nops.h> +#include <asm/nmi.h> -static unsigned char ftrace_nop[MCOUNT_INSN_SIZE]; +#ifdef CONFIG_DYNAMIC_FTRACE union ftrace_code_union { char code[MCOUNT_INSN_SIZE]; @@ -31,18 +34,12 @@ union ftrace_code_union { } __attribute__((packed)); }; - static int ftrace_calc_offset(long ip, long addr) { return (int)(addr - ip); } -unsigned char *ftrace_nop_replace(void) -{ - return ftrace_nop; -} - -unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr) +static unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr) { static union ftrace_code_union calc; @@ -56,7 +53,143 @@ unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr) return calc.code; } -int +/* + * Modifying code must take extra care. On an SMP machine, if + * the code being modified is also being executed on another CPU + * that CPU will have undefined results and possibly take a GPF. + * We use kstop_machine to stop other CPUS from exectuing code. + * But this does not stop NMIs from happening. We still need + * to protect against that. We separate out the modification of + * the code to take care of this. + * + * Two buffers are added: An IP buffer and a "code" buffer. + * + * 1) Put the instruction pointer into the IP buffer + * and the new code into the "code" buffer. + * 2) Set a flag that says we are modifying code + * 3) Wait for any running NMIs to finish. + * 4) Write the code + * 5) clear the flag. + * 6) Wait for any running NMIs to finish. + * + * If an NMI is executed, the first thing it does is to call + * "ftrace_nmi_enter". This will check if the flag is set to write + * and if it is, it will write what is in the IP and "code" buffers. + * + * The trick is, it does not matter if everyone is writing the same + * content to the code location. Also, if a CPU is executing code + * it is OK to write to that code location if the contents being written + * are the same as what exists. + */ + +static atomic_t in_nmi = ATOMIC_INIT(0); +static int mod_code_status; /* holds return value of text write */ +static int mod_code_write; /* set when NMI should do the write */ +static void *mod_code_ip; /* holds the IP to write to */ +static void *mod_code_newcode; /* holds the text to write to the IP */ + +static unsigned nmi_wait_count; +static atomic_t nmi_update_count = ATOMIC_INIT(0); + +int ftrace_arch_read_dyn_info(char *buf, int size) +{ + int r; + + r = snprintf(buf, size, "%u %u", + nmi_wait_count, + atomic_read(&nmi_update_count)); + return r; +} + +static void ftrace_mod_code(void) +{ + /* + * Yes, more than one CPU process can be writing to mod_code_status. + * (and the code itself) + * But if one were to fail, then they all should, and if one were + * to succeed, then they all should. + */ + mod_code_status = probe_kernel_write(mod_code_ip, mod_code_newcode, + MCOUNT_INSN_SIZE); + +} + +void ftrace_nmi_enter(void) +{ + atomic_inc(&in_nmi); + /* Must have in_nmi seen before reading write flag */ + smp_mb(); + if (mod_code_write) { + ftrace_mod_code(); + atomic_inc(&nmi_update_count); + } +} + +void ftrace_nmi_exit(void) +{ + /* Finish all executions before clearing in_nmi */ + smp_wmb(); + atomic_dec(&in_nmi); +} + +static void wait_for_nmi(void) +{ + int waited = 0; + + while (atomic_read(&in_nmi)) { + waited = 1; + cpu_relax(); + } + + if (waited) + nmi_wait_count++; +} + +static int +do_ftrace_mod_code(unsigned long ip, void *new_code) +{ + mod_code_ip = (void *)ip; + mod_code_newcode = new_code; + + /* The buffers need to be visible before we let NMIs write them */ + smp_wmb(); + + mod_code_write = 1; + + /* Make sure write bit is visible before we wait on NMIs */ + smp_mb(); + + wait_for_nmi(); + + /* Make sure all running NMIs have finished before we write the code */ + smp_mb(); + + ftrace_mod_code(); + + /* Make sure the write happens before clearing the bit */ + smp_wmb(); + + mod_code_write = 0; + + /* make sure NMIs see the cleared bit */ + smp_mb(); + + wait_for_nmi(); + + return mod_code_status; +} + + + + +static unsigned char ftrace_nop[MCOUNT_INSN_SIZE]; + +static unsigned char *ftrace_nop_replace(void) +{ + return ftrace_nop; +} + +static int ftrace_modify_code(unsigned long ip, unsigned char *old_code, unsigned char *new_code) { @@ -81,7 +214,7 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code, return -EINVAL; /* replace the text with the new text */ - if (probe_kernel_write((void *)ip, new_code, MCOUNT_INSN_SIZE)) + if (do_ftrace_mod_code(ip, new_code)) return -EPERM; sync_core(); @@ -89,6 +222,29 @@ ftrace_modify_code(unsigned long ip, unsigned char *old_code, return 0; } +int ftrace_make_nop(struct module *mod, + struct dyn_ftrace *rec, unsigned long addr) +{ + unsigned char *new, *old; + unsigned long ip = rec->ip; + + old = ftrace_call_replace(ip, addr); + new = ftrace_nop_replace(); + + return ftrace_modify_code(rec->ip, old, new); +} + +int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr) +{ + unsigned char *new, *old; + unsigned long ip = rec->ip; + + old = ftrace_nop_replace(); + new = ftrace_call_replace(ip, addr); + + return ftrace_modify_code(rec->ip, old, new); +} + int ftrace_update_ftrace_func(ftrace_func_t func) { unsigned long ip = (unsigned long)(&ftrace_call); @@ -165,3 +321,139 @@ int __init ftrace_dyn_arch_init(void *data) return 0; } +#endif + +#ifdef CONFIG_FUNCTION_RET_TRACER + +#ifndef CONFIG_DYNAMIC_FTRACE + +/* + * These functions are picked from those used on + * this page for dynamic ftrace. They have been + * simplified to ignore all traces in NMI context. + */ +static atomic_t in_nmi; + +void ftrace_nmi_enter(void) +{ + atomic_inc(&in_nmi); +} + +void ftrace_nmi_exit(void) +{ + atomic_dec(&in_nmi); +} +#endif /* !CONFIG_DYNAMIC_FTRACE */ + +/* Add a function return address to the trace stack on thread info.*/ +static int push_return_trace(unsigned long ret, unsigned long long time, + unsigned long func) +{ + int index; + + if (!current->ret_stack) + return -EBUSY; + + /* The return trace stack is full */ + if (current->curr_ret_stack == FTRACE_RETFUNC_DEPTH - 1) { + atomic_inc(¤t->trace_overrun); + return -EBUSY; + } + + index = ++current->curr_ret_stack; + barrier(); + current->ret_stack[index].ret = ret; + current->ret_stack[index].func = func; + current->ret_stack[index].calltime = time; + + return 0; +} + +/* Retrieve a function return address to the trace stack on thread info.*/ +static void pop_return_trace(unsigned long *ret, unsigned long long *time, + unsigned long *func, unsigned long *overrun) +{ + int index; + + index = current->curr_ret_stack; + *ret = current->ret_stack[index].ret; + *func = current->ret_stack[index].func; + *time = current->ret_stack[index].calltime; + *overrun = atomic_read(¤t->trace_overrun); + current->curr_ret_stack--; +} + +/* + * Send the trace to the ring-buffer. + * @return the original return address. + */ +unsigned long ftrace_return_to_handler(void) +{ + struct ftrace_retfunc trace; + pop_return_trace(&trace.ret, &trace.calltime, &trace.func, + &trace.overrun); + trace.rettime = cpu_clock(raw_smp_processor_id()); + ftrace_function_return(&trace); + + return trace.ret; +} + +/* + * Hook the return address and push it in the stack of return addrs + * in current thread info. + */ +void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr) +{ + unsigned long old; + unsigned long long calltime; + int faulted; + unsigned long return_hooker = (unsigned long) + &return_to_handler; + + /* Nmi's are currently unsupported */ + if (atomic_read(&in_nmi)) + return; + + /* + * Protect against fault, even if it shouldn't + * happen. This tool is too much intrusive to + * ignore such a protection. + */ + asm volatile( + "1: movl (%[parent_old]), %[old]\n" + "2: movl %[return_hooker], (%[parent_replaced])\n" + " movl $0, %[faulted]\n" + + ".section .fixup, \"ax\"\n" + "3: movl $1, %[faulted]\n" + ".previous\n" + + ".section __ex_table, \"a\"\n" + " .long 1b, 3b\n" + " .long 2b, 3b\n" + ".previous\n" + + : [parent_replaced] "=r" (parent), [old] "=r" (old), + [faulted] "=r" (faulted) + : [parent_old] "0" (parent), [return_hooker] "r" (return_hooker) + : "memory" + ); + + if (WARN_ON(faulted)) { + unregister_ftrace_return(); + return; + } + + if (WARN_ON(!__kernel_text_address(old))) { + unregister_ftrace_return(); + *parent = old; + return; + } + + calltime = cpu_clock(raw_smp_processor_id()); + + if (push_return_trace(old, calltime, self_addr) == -EBUSY) + *parent = old; +} + +#endif /* CONFIG_FUNCTION_RET_TRACER */ diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c index a03e7f6d90c..10786af9554 100644 --- a/arch/x86/kernel/stacktrace.c +++ b/arch/x86/kernel/stacktrace.c @@ -6,6 +6,7 @@ #include <linux/sched.h> #include <linux/stacktrace.h> #include <linux/module.h> +#include <linux/uaccess.h> #include <asm/stacktrace.h> static void save_stack_warning(void *data, char *msg) @@ -83,3 +84,66 @@ void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace) trace->entries[trace->nr_entries++] = ULONG_MAX; } EXPORT_SYMBOL_GPL(save_stack_trace_tsk); + +/* Userspace stacktrace - based on kernel/trace/trace_sysprof.c */ + +struct stack_frame { + const void __user *next_fp; + unsigned long ret_addr; +}; + +static int copy_stack_frame(const void __user *fp, struct stack_frame *frame) +{ + int ret; + + if (!access_ok(VERIFY_READ, fp, sizeof(*frame))) + return 0; + + ret = 1; + pagefault_disable(); + if (__copy_from_user_inatomic(frame, fp, sizeof(*frame))) + ret = 0; + pagefault_enable(); + + return ret; +} + +static inline void __save_stack_trace_user(struct stack_trace *trace) +{ + const struct pt_regs *regs = task_pt_regs(current); + const void __user *fp = (const void __user *)regs->bp; + + if (trace->nr_entries < trace->max_entries) + trace->entries[trace->nr_entries++] = regs->ip; + + while (trace->nr_entries < trace->max_entries) { + struct stack_frame frame; + + frame.next_fp = NULL; + frame.ret_addr = 0; + if (!copy_stack_frame(fp, &frame)) + break; + if ((unsigned long)fp < regs->sp) + break; + if (frame.ret_addr) { + trace->entries[trace->nr_entries++] = + frame.ret_addr; + } + if (fp == frame.next_fp) + break; + fp = frame.next_fp; + } +} + +void save_stack_trace_user(struct stack_trace *trace) +{ + /* + * Trace user stack if we are not a kernel thread + */ + if (current->mm) { + __save_stack_trace_user(trace); + } + if (trace->nr_entries < trace->max_entries) + trace->entries[trace->nr_entries++] = ULONG_MAX; +} + diff --git a/arch/x86/kernel/vsyscall_64.c b/arch/x86/kernel/vsyscall_64.c index 0b8b6690a86..6f3d3d4cd97 100644 --- a/arch/x86/kernel/vsyscall_64.c +++ b/arch/x86/kernel/vsyscall_64.c @@ -17,6 +17,9 @@ * want per guest time just set the kernel.vsyscall64 sysctl to 0. */ +/* Disable profiling for userspace code: */ +#define DISABLE_BRANCH_PROFILING + #include <linux/time.h> #include <linux/init.h> #include <linux/kernel.h> diff --git a/arch/x86/lib/usercopy_32.c b/arch/x86/lib/usercopy_32.c index 9e68075544f..4a20b2f9a38 100644 --- a/arch/x86/lib/usercopy_32.c +++ b/arch/x86/lib/usercopy_32.c @@ -39,7 +39,7 @@ static inline int __movsl_is_ok(unsigned long a1, unsigned long a2, unsigned lon #define __do_strncpy_from_user(dst, src, count, res) \ do { \ int __d0, __d1, __d2; \ - might_sleep(); \ + might_fault(); \ __asm__ __volatile__( \ " testl %1,%1\n" \ " jz 2f\n" \ @@ -126,7 +126,7 @@ EXPORT_SYMBOL(strncpy_from_user); #define __do_clear_user(addr,size) \ do { \ int __d0; \ - might_sleep(); \ + might_fault(); \ __asm__ __volatile__( \ "0: rep; stosl\n" \ " movl %2,%0\n" \ @@ -155,7 +155,7 @@ do { \ unsigned long clear_user(void __user *to, unsigned long n) { - might_sleep(); + might_fault(); if (access_ok(VERIFY_WRITE, to, n)) __do_clear_user(to, n); return n; @@ -197,7 +197,7 @@ long strnlen_user(const char __user *s, long n) unsigned long mask = -__addr_ok(s); unsigned long res, tmp; - might_sleep(); + might_fault(); __asm__ __volatile__( " testl %0, %0\n" diff --git a/arch/x86/lib/usercopy_64.c b/arch/x86/lib/usercopy_64.c index f4df6e7c718..64d6c84e635 100644 --- a/arch/x86/lib/usercopy_64.c +++ b/arch/x86/lib/usercopy_64.c @@ -15,7 +15,7 @@ #define __do_strncpy_from_user(dst,src,count,res) \ do { \ long __d0, __d1, __d2; \ - might_sleep(); \ + might_fault(); \ __asm__ __volatile__( \ " testq %1,%1\n" \ " jz 2f\n" \ @@ -64,7 +64,7 @@ EXPORT_SYMBOL(strncpy_from_user); unsigned long __clear_user(void __user *addr, unsigned long size) { long __d0; - might_sleep(); + might_fault(); /* no memory constraint because it doesn't change any memory gcc knows about */ asm volatile( diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile index fea4565ff57..d8cc96a2738 100644 --- a/arch/x86/mm/Makefile +++ b/arch/x86/mm/Makefile @@ -8,9 +8,8 @@ obj-$(CONFIG_X86_PTDUMP) += dump_pagetables.o obj-$(CONFIG_HIGHMEM) += highmem_32.o -obj-$(CONFIG_MMIOTRACE_HOOKS) += kmmio.o obj-$(CONFIG_MMIOTRACE) += mmiotrace.o -mmiotrace-y := pf_in.o mmio-mod.o +mmiotrace-y := kmmio.o pf_in.o mmio-mod.o obj-$(CONFIG_MMIOTRACE_TEST) += testmmiotrace.o obj-$(CONFIG_NUMA) += numa_$(BITS).o diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 31e8730fa24..4152d3c3b13 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -53,7 +53,7 @@ static inline int kmmio_fault(struct pt_regs *regs, unsigned long addr) { -#ifdef CONFIG_MMIOTRACE_HOOKS +#ifdef CONFIG_MMIOTRACE if (unlikely(is_kmmio_active())) if (kmmio_handler(regs, addr) == 1) return -1; diff --git a/arch/x86/vdso/vclock_gettime.c b/arch/x86/vdso/vclock_gettime.c index 1ef0f90813d..d9d35824c56 100644 --- a/arch/x86/vdso/vclock_gettime.c +++ b/arch/x86/vdso/vclock_gettime.c @@ -9,6 +9,9 @@ * Also alternative() doesn't work. */ +/* Disable profiling for userspace code: */ +#define DISABLE_BRANCH_PROFILING + #include <linux/kernel.h> #include <linux/posix-timers.h> #include <linux/time.h> diff --git a/drivers/char/sysrq.c b/drivers/char/sysrq.c index ce0d9da52a8..94966edfb44 100644 --- a/drivers/char/sysrq.c +++ b/drivers/char/sysrq.c @@ -274,6 +274,22 @@ static struct sysrq_key_op sysrq_showstate_blocked_op = { .enable_mask = SYSRQ_ENABLE_DUMP, }; +#ifdef CONFIG_TRACING +#include <linux/ftrace.h> + +static void sysrq_ftrace_dump(int key, struct tty_struct *tty) +{ + ftrace_dump(); +} +static struct sysrq_key_op sysrq_ftrace_dump_op = { + .handler = sysrq_ftrace_dump, + .help_msg = "dumpZ-ftrace-buffer", + .action_msg = "Dump ftrace buffer", + .enable_mask = SYSRQ_ENABLE_DUMP, +}; +#else +#define sysrq_ftrace_dump_op (*(struct sysrq_key_op *)0) +#endif static void sysrq_handle_showmem(int key, struct tty_struct *tty) { @@ -406,7 +422,7 @@ static struct sysrq_key_op *sysrq_key_table[36] = { NULL, /* x */ /* y: May be registered on sparc64 for global register dump */ NULL, /* y */ - NULL /* z */ + &sysrq_ftrace_dump_op, /* z */ }; /* key2index calculation, -1 on invalid index */ diff --git a/fs/seq_file.c b/fs/seq_file.c index eba2eabcd2b..f03220d7891 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -357,7 +357,18 @@ int seq_printf(struct seq_file *m, const char *f, ...) } EXPORT_SYMBOL(seq_printf); -static char *mangle_path(char *s, char *p, char *esc) +/** + * mangle_path - mangle and copy path to buffer beginning + * @s: buffer start + * @p: beginning of path in above buffer + * @esc: set of characters that need escaping + * + * Copy the path from @p to @s, replacing each occurrence of character from + * @esc with usual octal escape. + * Returns pointer past last written character in @s, or NULL in case of + * failure. + */ +char *mangle_path(char *s, char *p, char *esc) { while (s <= p) { char c = *p++; @@ -376,6 +387,7 @@ static char *mangle_path(char *s, char *p, char *esc) } return NULL; } +EXPORT_SYMBOL_GPL(mangle_path); /* * return the absolute path of 'dentry' residing in mount 'mnt'. diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 80744606bad..eba835a2c2c 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -45,6 +45,22 @@ #define MCOUNT_REC() #endif +#ifdef CONFIG_TRACE_BRANCH_PROFILING +#define LIKELY_PROFILE() VMLINUX_SYMBOL(__start_annotated_branch_profile) = .; \ + *(_ftrace_annotated_branch) \ + VMLINUX_SYMBOL(__stop_annotated_branch_profile) = .; +#else +#define LIKELY_PROFILE() +#endif + +#ifdef CONFIG_PROFILE_ALL_BRANCHES +#define BRANCH_PROFILE() VMLINUX_SYMBOL(__start_branch_profile) = .; \ + *(_ftrace_branch) \ + VMLINUX_SYMBOL(__stop_branch_profile) = .; +#else +#define BRANCH_PROFILE() +#endif + /* .data section */ #define DATA_DATA \ *(.data) \ @@ -60,9 +76,12 @@ VMLINUX_SYMBOL(__start___markers) = .; \ *(__markers) \ VMLINUX_SYMBOL(__stop___markers) = .; \ + . = ALIGN(32); \ VMLINUX_SYMBOL(__start___tracepoints) = .; \ *(__tracepoints) \ - VMLINUX_SYMBOL(__stop___tracepoints) = .; + VMLINUX_SYMBOL(__stop___tracepoints) = .; \ + LIKELY_PROFILE() \ + BRANCH_PROFILE() #define RO_DATA(align) \ . = ALIGN((align)); \ diff --git a/include/asm-m32r/system.h b/include/asm-m32r/system.h index 70a57c8c002..c980f5ba8de 100644 --- a/include/asm-m32r/system.h +++ b/include/asm-m32r/system.h @@ -23,7 +23,7 @@ */ #if defined(CONFIG_FRAME_POINTER) || \ - !defined(CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER) + !defined(CONFIG_SCHED_OMIT_FRAME_POINTER) #define M32R_PUSH_FP " push fp\n" #define M32R_POP_FP " pop fp\n" #else diff --git a/include/linux/compiler.h b/include/linux/compiler.h index 98115d9d04d..ea7c6be354b 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -59,8 +59,88 @@ extern void __chk_io_ptr(const volatile void __iomem *); * specific implementations come from the above header files */ -#define likely(x) __builtin_expect(!!(x), 1) -#define unlikely(x) __builtin_expect(!!(x), 0) +struct ftrace_branch_data { + const char *func; + const char *file; + unsigned line; + union { + struct { + unsigned long correct; + unsigned long incorrect; + }; + struct { + unsigned long miss; + unsigned long hit; + }; + }; +}; + +/* + * Note: DISABLE_BRANCH_PROFILING can be used by special lowlevel code + * to disable branch tracing on a per file basis. + */ +#if defined(CONFIG_TRACE_BRANCH_PROFILING) && !defined(DISABLE_BRANCH_PROFILING) +void ftrace_likely_update(struct ftrace_branch_data *f, int val, int expect); + +#define likely_notrace(x) __builtin_expect(!!(x), 1) +#define unlikely_notrace(x) __builtin_expect(!!(x), 0) + +#define __branch_check__(x, expect) ({ \ + int ______r; \ + static struct ftrace_branch_data \ + __attribute__((__aligned__(4))) \ + __attribute__((section("_ftrace_annotated_branch"))) \ + ______f = { \ + .func = __func__, \ + .file = __FILE__, \ + .line = __LINE__, \ + }; \ + ______r = likely_notrace(x); \ + ftrace_likely_update(&______f, ______r, expect); \ + ______r; \ + }) + +/* + * Using __builtin_constant_p(x) to ignore cases where the return + * value is always the same. This idea is taken from a similar patch + * written by Daniel Walker. + */ +# ifndef likely +# define likely(x) (__builtin_constant_p(x) ? !!(x) : __branch_check__(x, 1)) +# endif +# ifndef unlikely +# define unlikely(x) (__builtin_constant_p(x) ? !!(x) : __branch_check__(x, 0)) +# endif + +#ifdef CONFIG_PROFILE_ALL_BRANCHES +/* + * "Define 'is'", Bill Clinton + * "Define 'if'", Steven Rostedt + */ +#define if(cond) if (__builtin_constant_p((cond)) ? !!(cond) : \ + ({ \ + int ______r; \ + static struct ftrace_branch_data \ + __attribute__((__aligned__(4))) \ + __attribute__((section("_ftrace_branch"))) \ + ______f = { \ + .func = __func__, \ + .file = __FILE__, \ + .line = __LINE__, \ + }; \ + ______r = !!(cond); \ + if (______r) \ + ______f.hit++; \ + else \ + ______f.miss++; \ + ______r; \ + })) +#endif /* CONFIG_PROFILE_ALL_BRANCHES */ + +#else +# define likely(x) __builtin_expect(!!(x), 1) +# define unlikely(x) __builtin_expect(!!(x), 0) +#endif /* Optimization barrier */ #ifndef barrier diff --git a/include/linux/debug_locks.h b/include/linux/debug_locks.h index 4aaa4afb1cb..096476f1fb3 100644 --- a/include/linux/debug_locks.h +++ b/include/linux/debug_locks.h @@ -17,7 +17,7 @@ extern int debug_locks_off(void); ({ \ int __ret = 0; \ \ - if (unlikely(c)) { \ + if (!oops_in_progress && unlikely(c)) { \ if (debug_locks_off() && !debug_locks_silent) \ WARN_ON(1); \ __ret = 1; \ diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h index 703eb53cfa2..7854d87b97b 100644 --- a/include/linux/ftrace.h +++ b/include/linux/ftrace.h @@ -23,6 +23,45 @@ struct ftrace_ops { struct ftrace_ops *next; }; +extern int function_trace_stop; + +/* + * Type of the current tracing. + */ +enum ftrace_tracing_type_t { + FTRACE_TYPE_ENTER = 0, /* Hook the call of the function */ + FTRACE_TYPE_RETURN, /* Hook the return of the function */ +}; + +/* Current tracing type, default is FTRACE_TYPE_ENTER */ +extern enum ftrace_tracing_type_t ftrace_tracing_type; + +/** + * ftrace_stop - stop function tracer. + * + * A quick way to stop the function tracer. Note this an on off switch, + * it is not something that is recursive like preempt_disable. + * This does not disable the calling of mcount, it only stops the + * calling of functions from mcount. + */ +static inline void ftrace_stop(void) +{ + function_trace_stop = 1; +} + +/** + * ftrace_start - start the function tracer. + * + * This function is the inverse of ftrace_stop. This does not enable + * the function tracing if the function tracer is disabled. This only + * sets the function tracer flag to continue calling the functions + * from mcount. + */ +static inline void ftrace_start(void) +{ + function_trace_stop = 0; +} + /* * The ftrace_ops must be a static and should also * be read_mostly. These functions do modify read_mostly variables @@ -41,9 +80,13 @@ extern void ftrace_stub(unsigned long a0, unsigned long a1); # define unregister_ftrace_function(ops) do { } while (0) # define clear_ftrace_function(ops) do { } while (0) static inline void ftrace_kill(void) { } +static inline void ftrace_stop(void) { } +static inline void ftrace_start(void) { } #endif /* CONFIG_FUNCTION_TRACER */ #ifdef CONFIG_DYNAMIC_FTRACE +/* asm/ftrace.h must be defined for archs supporting dynamic ftrace */ +#include <asm/ftrace.h> enum { FTRACE_FL_FREE = (1 << 0), @@ -59,6 +102,7 @@ struct dyn_ftrace { struct list_head list; unsigned long ip; /* address of mcount call-site */ unsigned long flags; + struct dyn_arch_ftrace arch; }; int ftrace_force_update(void); @@ -66,19 +110,43 @@ void ftrace_set_filter(unsigned char *buf, int len, int reset); /* defined in arch */ extern int ftrace_ip_converted(unsigned long ip); -extern unsigned char *ftrace_nop_replace(void); -extern unsigned char *ftrace_call_replace(unsigned long ip, unsigned long addr); extern int ftrace_dyn_arch_init(void *data); extern int ftrace_update_ftrace_func(ftrace_func_t func); extern void ftrace_caller(void); extern void ftrace_call(void); extern void mcount_call(void); +#ifdef CONFIG_FUNCTION_RET_TRACER +extern void ftrace_return_caller(void); +#endif + +/** + * ftrace_make_nop - convert code into top + * @mod: module structure if called by module load initialization + * @rec: the mcount call site record + * @addr: the address that the call site should be calling + * + * This is a very sensitive operation and great care needs + * to be taken by the arch. The operation should carefully + * read the location, check to see if what is read is indeed + * what we expect it to be, and then on success of the compare, + * it should write to the location. + * + * The code segment at @rec->ip should be a caller to @addr + * + * Return must be: + * 0 on success + * -EFAULT on error reading the location + * -EINVAL on a failed compare of the contents + * -EPERM on error writing to the location + * Any other value will be considered a failure. + */ +extern int ftrace_make_nop(struct module *mod, + struct dyn_ftrace *rec, unsigned long addr); /** - * ftrace_modify_code - modify code segment - * @ip: the address of the code segment - * @old_code: the contents of what is expected to be there - * @new_code: the code to patch in + * ftrace_make_call - convert a nop call site into a call to addr + * @rec: the mcount call site record + * @addr: the address that the call site should call * * This is a very sensitive operation and great care needs * to be taken by the arch. The operation should carefully @@ -86,6 +154,8 @@ extern void mcount_call(void); * what we expect it to be, and then on success of the compare, * it should write to the location. * + * The code segment at @rec->ip should be a nop + * * Return must be: * 0 on success * -EFAULT on error reading the location @@ -93,8 +163,11 @@ extern void mcount_call(void); * -EPERM on error writing to the location * Any other value will be considered a failure. */ -extern int ftrace_modify_code(unsigned long ip, unsigned char *old_code, - unsigned char *new_code); +extern int ftrace_make_call(struct dyn_ftrace *rec, unsigned long addr); + + +/* May be defined in arch */ +extern int ftrace_arch_read_dyn_info(char *buf, int size); extern int skip_trace(unsigned long ip); @@ -102,7 +175,6 @@ extern void ftrace_release(void *start, unsigned long size); extern void ftrace_disable_daemon(void); extern void ftrace_enable_daemon(void); - #else # define skip_trace(ip) ({ 0; }) # define ftrace_force_update() ({ 0; }) @@ -181,6 +253,12 @@ static inline void __ftrace_enabled_restore(int enabled) #endif #ifdef CONFIG_TRACING +extern int ftrace_dump_on_oops; + +extern void tracing_start(void); +extern void tracing_stop(void); +extern void ftrace_off_permanent(void); + extern void ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3); @@ -211,6 +289,9 @@ ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3) { } static inline int ftrace_printk(const char *fmt, ...) __attribute__ ((format (printf, 1, 0))); +static inline void tracing_start(void) { } +static inline void tracing_stop(void) { } +static inline void ftrace_off_permanent(void) { } static inline int ftrace_printk(const char *fmt, ...) { @@ -221,33 +302,44 @@ static inline void ftrace_dump(void) { } #ifdef CONFIG_FTRACE_MCOUNT_RECORD extern void ftrace_init(void); -extern void ftrace_init_module(unsigned long *start, unsigned long *end); +extern void ftrace_init_module(struct module *mod, + unsigned long *start, unsigned long *end); #else static inline void ftrace_init(void) { } static inline void -ftrace_init_module(unsigned long *start, unsigned long *end) { } +ftrace_init_module(struct module *mod, + unsigned long *start, unsigned long *end) { } #endif -struct boot_trace { - pid_t caller; - char func[KSYM_NAME_LEN]; - int result; - unsigned long long duration; /* usecs */ - ktime_t calltime; - ktime_t rettime; +/* + * Structure that defines a return function trace. + */ +struct ftrace_retfunc { + unsigned long ret; /* Return address */ + unsigned long func; /* Current function */ + unsigned long long calltime; + unsigned long long rettime; + /* Number of functions that overran the depth limit for current task */ + unsigned long overrun; }; -#ifdef CONFIG_BOOT_TRACER -extern void trace_boot(struct boot_trace *it, initcall_t fn); -extern void start_boot_trace(void); -extern void stop_boot_trace(void); -#else -static inline void trace_boot(struct boot_trace *it, initcall_t fn) { } -static inline void start_boot_trace(void) { } -static inline void stop_boot_trace(void) { } -#endif +#ifdef CONFIG_FUNCTION_RET_TRACER +#define FTRACE_RETFUNC_DEPTH 50 +#define FTRACE_RETSTACK_ALLOC_SIZE 32 +/* Type of a callback handler of tracing return function */ +typedef void (*trace_function_return_t)(struct ftrace_retfunc *); +extern int register_ftrace_return(trace_function_return_t func); +/* The current handler in use */ +extern trace_function_return_t ftrace_function_return; +extern void unregister_ftrace_return(void); +extern void ftrace_retfunc_init_task(struct task_struct *t); +extern void ftrace_retfunc_exit_task(struct task_struct *t); +#else +static inline void ftrace_retfunc_init_task(struct task_struct *t) { } +static inline void ftrace_retfunc_exit_task(struct task_struct *t) { } +#endif #endif /* _LINUX_FTRACE_H */ diff --git a/include/linux/ftrace_irq.h b/include/linux/ftrace_irq.h new file mode 100644 index 00000000000..0b4df55d7a7 --- /dev/null +++ b/include/linux/ftrace_irq.h @@ -0,0 +1,13 @@ +#ifndef _LINUX_FTRACE_IRQ_H +#define _LINUX_FTRACE_IRQ_H + + +#if defined(CONFIG_DYNAMIC_FTRACE) || defined(CONFIG_FUNCTION_RET_TRACER) +extern void ftrace_nmi_enter(void); +extern void ftrace_nmi_exit(void); +#else +static inline void ftrace_nmi_enter(void) { } +static inline void ftrace_nmi_exit(void) { } +#endif + +#endif /* _LINUX_FTRACE_IRQ_H */ diff --git a/include/linux/futex.h b/include/linux/futex.h index 586ab56a3ec..8f627b9ae2b 100644 --- a/include/linux/futex.h +++ b/include/linux/futex.h @@ -164,6 +164,8 @@ union futex_key { } both; }; +#define FUTEX_KEY_INIT (union futex_key) { .both = { .ptr = NULL } } + #ifdef CONFIG_FUTEX extern void exit_robust_list(struct task_struct *curr); extern void exit_pi_state_list(struct task_struct *curr); diff --git a/include/linux/hardirq.h b/include/linux/hardirq.h index 181006cc94a..89a56d79e4c 100644 --- a/include/linux/hardirq.h +++ b/include/linux/hardirq.h @@ -4,6 +4,7 @@ #include <linux/preempt.h> #include <linux/smp_lock.h> #include <linux/lockdep.h> +#include <linux/ftrace_irq.h> #include <asm/hardirq.h> #include <asm/system.h> @@ -161,7 +162,17 @@ extern void irq_enter(void); */ extern void irq_exit(void); -#define nmi_enter() do { lockdep_off(); __irq_enter(); } while (0) -#define nmi_exit() do { __irq_exit(); lockdep_on(); } while (0) +#define nmi_enter() \ + do { \ + ftrace_nmi_enter(); \ + lockdep_off(); \ + __irq_enter(); \ + } while (0) +#define nmi_exit() \ + do { \ + __irq_exit(); \ + lockdep_on(); \ + ftrace_nmi_exit(); \ + } while (0) #endif /* LINUX_HARDIRQ_H */ diff --git a/include/linux/kernel.h b/include/linux/kernel.h index dc7e0d0a647..269df5a17b3 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -141,6 +141,15 @@ extern int _cond_resched(void); (__x < 0) ? -__x : __x; \ }) +#ifdef CONFIG_PROVE_LOCKING +void might_fault(void); +#else +static inline void might_fault(void) +{ + might_sleep(); +} +#endif + extern struct atomic_notifier_head panic_notifier_list; extern long (*panic_blink)(long time); NORET_TYPE void panic(const char * fmt, ...) @@ -188,6 +197,8 @@ extern unsigned long long memparse(const char *ptr, char **retptr); extern int core_kernel_text(unsigned long addr); extern int __kernel_text_address(unsigned long addr); extern int kernel_text_address(unsigned long addr); +extern int func_ptr_is_kernel_text(void *ptr); + struct pid; extern struct pid *session_of_pgrp(struct pid *pgrp); diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h index 29aec6e1002..8956daf64ab 100644 --- a/include/linux/lockdep.h +++ b/include/linux/lockdep.h @@ -73,6 +73,8 @@ struct lock_class_key { struct lockdep_subclass_key subkeys[MAX_LOCKDEP_SUBCLASSES]; }; +#define LOCKSTAT_POINTS 4 + /* * The lock-class itself: */ @@ -119,7 +121,8 @@ struct lock_class { int name_version; #ifdef CONFIG_LOCK_STAT - unsigned long contention_point[4]; + unsigned long contention_point[LOCKSTAT_POINTS]; + unsigned long contending_point[LOCKSTAT_POINTS]; #endif }; @@ -144,6 +147,7 @@ enum bounce_type { struct lock_class_stats { unsigned long contention_point[4]; + unsigned long contending_point[4]; struct lock_time read_waittime; struct lock_time write_waittime; struct lock_time read_holdtime; @@ -165,6 +169,7 @@ struct lockdep_map { const char *name; #ifdef CONFIG_LOCK_STAT int cpu; + unsigned long ip; #endif }; @@ -356,7 +361,7 @@ struct lock_class_key { }; #ifdef CONFIG_LOCK_STAT extern void lock_contended(struct lockdep_map *lock, unsigned long ip); -extern void lock_acquired(struct lockdep_map *lock); +extern void lock_acquired(struct lockdep_map *lock, unsigned long ip); #define LOCK_CONTENDED(_lock, try, lock) \ do { \ @@ -364,13 +369,13 @@ do { \ lock_contended(&(_lock)->dep_map, _RET_IP_); \ lock(_lock); \ } \ - lock_acquired(&(_lock)->dep_map); \ + lock_acquired(&(_lock)->dep_map, _RET_IP_); \ } while (0) #else /* CONFIG_LOCK_STAT */ #define lock_contended(lockdep_map, ip) do {} while (0) -#define lock_acquired(lockdep_map) do {} while (0) +#define lock_acquired(lockdep_map, ip) do {} while (0) #define LOCK_CONTENDED(_lock, try, lock) \ lock(_lock) @@ -481,4 +486,22 @@ static inline void print_irqtrace_events(struct task_struct *curr) # define lock_map_release(l) do { } while (0) #endif +#ifdef CONFIG_PROVE_LOCKING +# define might_lock(lock) \ +do { \ + typecheck(struct lockdep_map *, &(lock)->dep_map); \ + lock_acquire(&(lock)->dep_map, 0, 0, 0, 2, NULL, _THIS_IP_); \ + lock_release(&(lock)->dep_map, 0, _THIS_IP_); \ +} while (0) +# define might_lock_read(lock) \ +do { \ + typecheck(struct lockdep_map *, &(lock)->dep_map); \ + lock_acquire(&(lock)->dep_map, 0, 0, 1, 2, NULL, _THIS_IP_); \ + lock_release(&(lock)->dep_map, 0, _THIS_IP_); \ +} while (0) +#else +# define might_lock(lock) do { } while (0) +# define might_lock_read(lock) do { } while (0) +#endif + #endif /* __LINUX_LOCKDEP_H */ diff --git a/include/linux/marker.h b/include/linux/marker.h index 889196c7fbb..34c14bc957f 100644 --- a/include/linux/marker.h +++ b/include/linux/marker.h @@ -12,6 +12,7 @@ * See the file COPYING for more details. */ +#include <stdarg.h> #include <linux/types.h> struct module; @@ -48,10 +49,28 @@ struct marker { void (*call)(const struct marker *mdata, void *call_private, ...); struct marker_probe_closure single; struct marker_probe_closure *multi; + const char *tp_name; /* Optional tracepoint name */ + void *tp_cb; /* Optional tracepoint callback */ } __attribute__((aligned(8))); #ifdef CONFIG_MARKERS +#define _DEFINE_MARKER(name, tp_name_str, tp_cb, format) \ + static const char __mstrtab_##name[] \ + __attribute__((section("__markers_strings"))) \ + = #name "\0" format; \ + static struct marker __mark_##name \ + __attribute__((section("__markers"), aligned(8))) = \ + { __mstrtab_##name, &__mstrtab_##name[sizeof(#name)], \ + 0, 0, marker_probe_cb, { __mark_empty_function, NULL},\ + NULL, tp_name_str, tp_cb } + +#define DEFINE_MARKER(name, format) \ + _DEFINE_MARKER(name, NULL, NULL, format) + +#define DEFINE_MARKER_TP(name, tp_name, tp_cb, format) \ + _DEFINE_MARKER(name, #tp_name, tp_cb, format) + /* * Note : the empty asm volatile with read constraint is used here instead of a * "used" attribute to fix a gcc 4.1.x bug. @@ -65,14 +84,7 @@ struct marker { */ #define __trace_mark(generic, name, call_private, format, args...) \ do { \ - static const char __mstrtab_##name[] \ - __attribute__((section("__markers_strings"))) \ - = #name "\0" format; \ - static struct marker __mark_##name \ - __attribute__((section("__markers"), aligned(8))) = \ - { __mstrtab_##name, &__mstrtab_##name[sizeof(#name)], \ - 0, 0, marker_probe_cb, \ - { __mark_empty_function, NULL}, NULL }; \ + DEFINE_MARKER(name, format); \ __mark_check_format(format, ## args); \ if (unlikely(__mark_##name.state)) { \ (*__mark_##name.call) \ @@ -80,14 +92,39 @@ struct marker { } \ } while (0) +#define __trace_mark_tp(name, call_private, tp_name, tp_cb, format, args...) \ + do { \ + void __check_tp_type(void) \ + { \ + register_trace_##tp_name(tp_cb); \ + } \ + DEFINE_MARKER_TP(name, tp_name, tp_cb, format); \ + __mark_check_format(format, ## args); \ + (*__mark_##name.call)(&__mark_##name, call_private, \ + ## args); \ + } while (0) + extern void marker_update_probe_range(struct marker *begin, struct marker *end); + +#define GET_MARKER(name) (__mark_##name) + #else /* !CONFIG_MARKERS */ +#define DEFINE_MARKER(name, tp_name, tp_cb, format) #define __trace_mark(generic, name, call_private, format, args...) \ __mark_check_format(format, ## args) +#define __trace_mark_tp(name, call_private, tp_name, tp_cb, format, args...) \ + do { \ + void __check_tp_type(void) \ + { \ + register_trace_##tp_name(tp_cb); \ + } \ + __mark_check_format(format, ## args); \ + } while (0) static inline void marker_update_probe_range(struct marker *begin, struct marker *end) { } +#define GET_MARKER(name) #endif /* CONFIG_MARKERS */ /** @@ -117,6 +154,20 @@ static inline void marker_update_probe_range(struct marker *begin, __trace_mark(1, name, NULL, format, ## args) /** + * trace_mark_tp - Marker in a tracepoint callback + * @name: marker name, not quoted. + * @tp_name: tracepoint name, not quoted. + * @tp_cb: tracepoint callback. Should have an associated global symbol so it + * is not optimized away by the compiler (should not be static). + * @format: format string + * @args...: variable argument list + * + * Places a marker in a tracepoint callback. + */ +#define trace_mark_tp(name, tp_name, tp_cb, format, args...) \ + __trace_mark_tp(name, NULL, tp_name, tp_cb, format, ## args) + +/** * MARK_NOARGS - Format string for a marker with no argument. */ #define MARK_NOARGS " " @@ -136,8 +187,6 @@ extern marker_probe_func __mark_empty_function; extern void marker_probe_cb(const struct marker *mdata, void *call_private, ...); -extern void marker_probe_cb_noarg(const struct marker *mdata, - void *call_private, ...); /* * Connect a probe to a marker. diff --git a/include/linux/mutex.h b/include/linux/mutex.h index bc6da10ceee..7a0e5c4f807 100644 --- a/include/linux/mutex.h +++ b/include/linux/mutex.h @@ -144,6 +144,8 @@ extern int __must_check mutex_lock_killable(struct mutex *lock); /* * NOTE: mutex_trylock() follows the spin_trylock() convention, * not the down_trylock() convention! + * + * Returns 1 if the mutex has been acquired successfully, and 0 on contention. */ extern int mutex_trylock(struct mutex *lock); extern void mutex_unlock(struct mutex *lock); diff --git a/include/linux/rcuclassic.h b/include/linux/rcuclassic.h index 5f89b62e698..301dda829e3 100644 --- a/include/linux/rcuclassic.h +++ b/include/linux/rcuclassic.h @@ -41,7 +41,7 @@ #include <linux/seqlock.h> #ifdef CONFIG_RCU_CPU_STALL_DETECTOR -#define RCU_SECONDS_TILL_STALL_CHECK ( 3 * HZ) /* for rcp->jiffies_stall */ +#define RCU_SECONDS_TILL_STALL_CHECK (10 * HZ) /* for rcp->jiffies_stall */ #define RCU_SECONDS_TILL_STALL_RECHECK (30 * HZ) /* for rcp->jiffies_stall */ #endif /* #ifdef CONFIG_RCU_CPU_STALL_DETECTOR */ diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h index 86f1f5e43e3..895dc9c1088 100644 --- a/include/linux/rcupdate.h +++ b/include/linux/rcupdate.h @@ -142,6 +142,7 @@ struct rcu_head { * on the write-side to insure proper synchronization. */ #define rcu_read_lock_sched() preempt_disable() +#define rcu_read_lock_sched_notrace() preempt_disable_notrace() /* * rcu_read_unlock_sched - marks the end of a RCU-classic critical section @@ -149,6 +150,7 @@ struct rcu_head { * See rcu_read_lock_sched for more information. */ #define rcu_read_unlock_sched() preempt_enable() +#define rcu_read_unlock_sched_notrace() preempt_enable_notrace() diff --git a/include/linux/ring_buffer.h b/include/linux/ring_buffer.h index e097c2e6b6d..3bb87a753fa 100644 --- a/include/linux/ring_buffer.h +++ b/include/linux/ring_buffer.h @@ -122,6 +122,7 @@ void ring_buffer_normalize_time_stamp(int cpu, u64 *ts); void tracing_on(void); void tracing_off(void); +void tracing_off_permanent(void); enum ring_buffer_flags { RB_FL_OVERWRITE = 1 << 0, diff --git a/include/linux/sched.h b/include/linux/sched.h index 55e30d11447..bf953932e67 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -249,7 +249,7 @@ extern void init_idle_bootup_task(struct task_struct *idle); extern int runqueue_is_locked(void); extern void task_rq_unlock_wait(struct task_struct *p); -extern cpumask_t nohz_cpu_mask; +extern cpumask_var_t nohz_cpu_mask; #if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ) extern int select_nohz_load_balancer(int cpu); #else @@ -259,8 +259,6 @@ static inline int select_nohz_load_balancer(int cpu) } #endif -extern unsigned long rt_needs_cpu(int cpu); - /* * Only dump TASK_* tasks. (0 for all tasks) */ @@ -777,7 +775,6 @@ enum cpu_idle_type { struct sched_group { struct sched_group *next; /* Must be a circular list */ - cpumask_t cpumask; /* * CPU power of this group, SCHED_LOAD_SCALE being max power for a @@ -790,8 +787,15 @@ struct sched_group { * (see include/linux/reciprocal_div.h) */ u32 reciprocal_cpu_power; + + unsigned long cpumask[]; }; +static inline struct cpumask *sched_group_cpus(struct sched_group *sg) +{ + return to_cpumask(sg->cpumask); +} + enum sched_domain_level { SD_LV_NONE = 0, SD_LV_SIBLING, @@ -815,7 +819,6 @@ struct sched_domain { struct sched_domain *parent; /* top domain must be null terminated */ struct sched_domain *child; /* bottom domain must be null terminated */ struct sched_group *groups; /* the balancing groups of the domain */ - cpumask_t span; /* span of all CPUs in this domain */ unsigned long min_interval; /* Minimum balance interval ms */ unsigned long max_interval; /* Maximum balance interval ms */ unsigned int busy_factor; /* less balancing by factor if busy */ @@ -870,9 +873,17 @@ struct sched_domain { #ifdef CONFIG_SCHED_DEBUG char *name; #endif + + /* span of all CPUs in this domain */ + unsigned long span[]; }; -extern void partition_sched_domains(int ndoms_new, cpumask_t *doms_new, +static inline struct cpumask *sched_domain_span(struct sched_domain *sd) +{ + return to_cpumask(sd->span); +} + +extern void partition_sched_domains(int ndoms_new, struct cpumask *doms_new, struct sched_domain_attr *dattr_new); extern int arch_reinit_sched_domains(void); @@ -881,7 +892,7 @@ extern int arch_reinit_sched_domains(void); struct sched_domain_attr; static inline void -partition_sched_domains(int ndoms_new, cpumask_t *doms_new, +partition_sched_domains(int ndoms_new, struct cpumask *doms_new, struct sched_domain_attr *dattr_new) { } @@ -963,7 +974,7 @@ struct sched_class { void (*task_wake_up) (struct rq *this_rq, struct task_struct *task); void (*set_cpus_allowed)(struct task_struct *p, - const cpumask_t *newmask); + const struct cpumask *newmask); void (*rq_online)(struct rq *rq); void (*rq_offline)(struct rq *rq); @@ -1356,6 +1367,17 @@ struct task_struct { unsigned long default_timer_slack_ns; struct list_head *scm_work_list; +#ifdef CONFIG_FUNCTION_RET_TRACER + /* Index of current stored adress in ret_stack */ + int curr_ret_stack; + /* Stack of return addresses for return function tracing */ + struct ftrace_ret_stack *ret_stack; + /* + * Number of functions that haven't been traced + * because of depth overrun. + */ + atomic_t trace_overrun; +#endif }; /* @@ -1594,12 +1616,12 @@ extern cputime_t task_gtime(struct task_struct *p); #ifdef CONFIG_SMP extern int set_cpus_allowed_ptr(struct task_struct *p, - const cpumask_t *new_mask); + const struct cpumask *new_mask); #else static inline int set_cpus_allowed_ptr(struct task_struct *p, - const cpumask_t *new_mask) + const struct cpumask *new_mask) { - if (!cpu_isset(0, *new_mask)) + if (!cpumask_test_cpu(0, new_mask)) return -EINVAL; return 0; } @@ -2212,8 +2234,8 @@ __trace_special(void *__tr, void *__data, } #endif -extern long sched_setaffinity(pid_t pid, const cpumask_t *new_mask); -extern long sched_getaffinity(pid_t pid, cpumask_t *mask); +extern long sched_setaffinity(pid_t pid, const struct cpumask *new_mask); +extern long sched_getaffinity(pid_t pid, struct cpumask *mask); extern int sched_mc_power_savings, sched_smt_power_savings; diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h index dc50bcc282a..b3dfa72f13b 100644 --- a/include/linux/seq_file.h +++ b/include/linux/seq_file.h @@ -34,6 +34,7 @@ struct seq_operations { #define SEQ_SKIP 1 +char *mangle_path(char *s, char *p, char *esc); int seq_open(struct file *, const struct seq_operations *); ssize_t seq_read(struct file *, char __user *, size_t, loff_t *); loff_t seq_lseek(struct file *, loff_t, int); diff --git a/include/linux/stacktrace.h b/include/linux/stacktrace.h index b106fd8e0d5..1a8cecc4f38 100644 --- a/include/linux/stacktrace.h +++ b/include/linux/stacktrace.h @@ -15,9 +15,17 @@ extern void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace); extern void print_stack_trace(struct stack_trace *trace, int spaces); + +#ifdef CONFIG_USER_STACKTRACE_SUPPORT +extern void save_stack_trace_user(struct stack_trace *trace); +#else +# define save_stack_trace_user(trace) do { } while (0) +#endif + #else # define save_stack_trace(trace) do { } while (0) # define save_stack_trace_tsk(tsk, trace) do { } while (0) +# define save_stack_trace_user(trace) do { } while (0) # define print_stack_trace(trace, spaces) do { } while (0) #endif diff --git a/include/linux/tracepoint.h b/include/linux/tracepoint.h index c5bb39c7a77..75700545836 100644 --- a/include/linux/tracepoint.h +++ b/include/linux/tracepoint.h @@ -24,8 +24,12 @@ struct tracepoint { const char *name; /* Tracepoint name */ int state; /* State. */ void **funcs; -} __attribute__((aligned(8))); - +} __attribute__((aligned(32))); /* + * Aligned on 32 bytes because it is + * globally visible and gcc happily + * align these on the structure size. + * Keep in sync with vmlinux.lds.h. + */ #define TPPROTO(args...) args #define TPARGS(args...) args @@ -40,14 +44,14 @@ struct tracepoint { do { \ void **it_func; \ \ - rcu_read_lock_sched(); \ + rcu_read_lock_sched_notrace(); \ it_func = rcu_dereference((tp)->funcs); \ if (it_func) { \ do { \ ((void(*)(proto))(*it_func))(args); \ } while (*(++it_func)); \ } \ - rcu_read_unlock_sched(); \ + rcu_read_unlock_sched_notrace(); \ } while (0) /* @@ -55,35 +59,40 @@ struct tracepoint { * not add unwanted padding between the beginning of the section and the * structure. Force alignment to the same alignment as the section start. */ -#define DEFINE_TRACE(name, proto, args) \ +#define DECLARE_TRACE(name, proto, args) \ + extern struct tracepoint __tracepoint_##name; \ static inline void trace_##name(proto) \ { \ - static const char __tpstrtab_##name[] \ - __attribute__((section("__tracepoints_strings"))) \ - = #name ":" #proto; \ - static struct tracepoint __tracepoint_##name \ - __attribute__((section("__tracepoints"), aligned(8))) = \ - { __tpstrtab_##name, 0, NULL }; \ if (unlikely(__tracepoint_##name.state)) \ __DO_TRACE(&__tracepoint_##name, \ TPPROTO(proto), TPARGS(args)); \ } \ static inline int register_trace_##name(void (*probe)(proto)) \ { \ - return tracepoint_probe_register(#name ":" #proto, \ - (void *)probe); \ + return tracepoint_probe_register(#name, (void *)probe); \ } \ - static inline void unregister_trace_##name(void (*probe)(proto))\ + static inline int unregister_trace_##name(void (*probe)(proto)) \ { \ - tracepoint_probe_unregister(#name ":" #proto, \ - (void *)probe); \ + return tracepoint_probe_unregister(#name, (void *)probe);\ } +#define DEFINE_TRACE(name) \ + static const char __tpstrtab_##name[] \ + __attribute__((section("__tracepoints_strings"))) = #name; \ + struct tracepoint __tracepoint_##name \ + __attribute__((section("__tracepoints"), aligned(32))) = \ + { __tpstrtab_##name, 0, NULL } + +#define EXPORT_TRACEPOINT_SYMBOL_GPL(name) \ + EXPORT_SYMBOL_GPL(__tracepoint_##name) +#define EXPORT_TRACEPOINT_SYMBOL(name) \ + EXPORT_SYMBOL(__tracepoint_##name) + extern void tracepoint_update_probe_range(struct tracepoint *begin, struct tracepoint *end); #else /* !CONFIG_TRACEPOINTS */ -#define DEFINE_TRACE(name, proto, args) \ +#define DECLARE_TRACE(name, proto, args) \ static inline void _do_trace_##name(struct tracepoint *tp, proto) \ { } \ static inline void trace_##name(proto) \ @@ -92,8 +101,14 @@ extern void tracepoint_update_probe_range(struct tracepoint *begin, { \ return -ENOSYS; \ } \ - static inline void unregister_trace_##name(void (*probe)(proto))\ - { } + static inline int unregister_trace_##name(void (*probe)(proto)) \ + { \ + return -ENOSYS; \ + } + +#define DEFINE_TRACE(name) +#define EXPORT_TRACEPOINT_SYMBOL_GPL(name) +#define EXPORT_TRACEPOINT_SYMBOL(name) static inline void tracepoint_update_probe_range(struct tracepoint *begin, struct tracepoint *end) @@ -112,6 +127,10 @@ extern int tracepoint_probe_register(const char *name, void *probe); */ extern int tracepoint_probe_unregister(const char *name, void *probe); +extern int tracepoint_probe_register_noupdate(const char *name, void *probe); +extern int tracepoint_probe_unregister_noupdate(const char *name, void *probe); +extern void tracepoint_probe_update_all(void); + struct tracepoint_iter { struct module *module; struct tracepoint *tracepoint; diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h index fec6decfb98..6b58367d145 100644 --- a/include/linux/uaccess.h +++ b/include/linux/uaccess.h @@ -78,7 +78,7 @@ static inline unsigned long __copy_from_user_nocache(void *to, \ set_fs(KERNEL_DS); \ pagefault_disable(); \ - ret = __get_user(retval, (__force typeof(retval) __user *)(addr)); \ + ret = __copy_from_user_inatomic(&(retval), (__force typeof(retval) __user *)(addr), sizeof(retval)); \ pagefault_enable(); \ set_fs(old_fs); \ ret; \ diff --git a/include/trace/boot.h b/include/trace/boot.h new file mode 100644 index 00000000000..6b54537eab0 --- /dev/null +++ b/include/trace/boot.h @@ -0,0 +1,56 @@ +#ifndef _LINUX_TRACE_BOOT_H +#define _LINUX_TRACE_BOOT_H + +/* + * Structure which defines the trace of an initcall + * while it is called. + * You don't have to fill the func field since it is + * only used internally by the tracer. + */ +struct boot_trace_call { + pid_t caller; + char func[KSYM_NAME_LEN]; +}; + +/* + * Structure which defines the trace of an initcall + * while it returns. + */ +struct boot_trace_ret { + char func[KSYM_NAME_LEN]; + int result; + unsigned long long duration; /* nsecs */ +}; + +#ifdef CONFIG_BOOT_TRACER +/* Append the traces on the ring-buffer */ +extern void trace_boot_call(struct boot_trace_call *bt, initcall_t fn); +extern void trace_boot_ret(struct boot_trace_ret *bt, initcall_t fn); + +/* Tells the tracer that smp_pre_initcall is finished. + * So we can start the tracing + */ +extern void start_boot_trace(void); + +/* Resume the tracing of other necessary events + * such as sched switches + */ +extern void enable_boot_trace(void); + +/* Suspend this tracing. Actually, only sched_switches tracing have + * to be suspended. Initcalls doesn't need it.) + */ +extern void disable_boot_trace(void); +#else +static inline +void trace_boot_call(struct boot_trace_call *bt, initcall_t fn) { } + +static inline +void trace_boot_ret(struct boot_trace_ret *bt, initcall_t fn) { } + +static inline void start_boot_trace(void) { } +static inline void enable_boot_trace(void) { } +static inline void disable_boot_trace(void) { } +#endif /* CONFIG_BOOT_TRACER */ + +#endif /* __LINUX_TRACE_BOOT_H */ diff --git a/include/trace/sched.h b/include/trace/sched.h index ad47369d01b..9b2854abf7e 100644 --- a/include/trace/sched.h +++ b/include/trace/sched.h @@ -4,52 +4,52 @@ #include <linux/sched.h> #include <linux/tracepoint.h> -DEFINE_TRACE(sched_kthread_stop, +DECLARE_TRACE(sched_kthread_stop, TPPROTO(struct task_struct *t), TPARGS(t)); -DEFINE_TRACE(sched_kthread_stop_ret, +DECLARE_TRACE(sched_kthread_stop_ret, TPPROTO(int ret), TPARGS(ret)); -DEFINE_TRACE(sched_wait_task, +DECLARE_TRACE(sched_wait_task, TPPROTO(struct rq *rq, struct task_struct *p), TPARGS(rq, p)); -DEFINE_TRACE(sched_wakeup, +DECLARE_TRACE(sched_wakeup, TPPROTO(struct rq *rq, struct task_struct *p), TPARGS(rq, p)); -DEFINE_TRACE(sched_wakeup_new, +DECLARE_TRACE(sched_wakeup_new, TPPROTO(struct rq *rq, struct task_struct *p), TPARGS(rq, p)); -DEFINE_TRACE(sched_switch, +DECLARE_TRACE(sched_switch, TPPROTO(struct rq *rq, struct task_struct *prev, struct task_struct *next), TPARGS(rq, prev, next)); -DEFINE_TRACE(sched_migrate_task, +DECLARE_TRACE(sched_migrate_task, TPPROTO(struct rq *rq, struct task_struct *p, int dest_cpu), TPARGS(rq, p, dest_cpu)); -DEFINE_TRACE(sched_process_free, +DECLARE_TRACE(sched_process_free, TPPROTO(struct task_struct *p), TPARGS(p)); -DEFINE_TRACE(sched_process_exit, +DECLARE_TRACE(sched_process_exit, TPPROTO(struct task_struct *p), TPARGS(p)); -DEFINE_TRACE(sched_process_wait, +DECLARE_TRACE(sched_process_wait, TPPROTO(struct pid *pid), TPARGS(pid)); -DEFINE_TRACE(sched_process_fork, +DECLARE_TRACE(sched_process_fork, TPPROTO(struct task_struct *parent, struct task_struct *child), TPARGS(parent, child)); -DEFINE_TRACE(sched_signal_send, +DECLARE_TRACE(sched_signal_send, TPPROTO(int sig, struct task_struct *p), TPARGS(sig, p)); diff --git a/init/Kconfig b/init/Kconfig index f763762d544..f291f086caa 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -808,6 +808,7 @@ config TRACEPOINTS config MARKERS bool "Activate markers" + depends on TRACEPOINTS help Place an empty function call at each marker site. Can be dynamically changed for a probe function. diff --git a/init/main.c b/init/main.c index 7e117a231af..79213c0785d 100644 --- a/init/main.c +++ b/init/main.c @@ -63,6 +63,7 @@ #include <linux/signal.h> #include <linux/idr.h> #include <linux/ftrace.h> +#include <trace/boot.h> #include <asm/io.h> #include <asm/bugs.h> @@ -703,31 +704,35 @@ core_param(initcall_debug, initcall_debug, bool, 0644); int do_one_initcall(initcall_t fn) { int count = preempt_count(); - ktime_t delta; + ktime_t calltime, delta, rettime; char msgbuf[64]; - struct boot_trace it; + struct boot_trace_call call; + struct boot_trace_ret ret; if (initcall_debug) { - it.caller = task_pid_nr(current); - printk("calling %pF @ %i\n", fn, it.caller); - it.calltime = ktime_get(); + call.caller = task_pid_nr(current); + printk("calling %pF @ %i\n", fn, call.caller); + calltime = ktime_get(); + trace_boot_call(&call, fn); + enable_boot_trace(); } - it.result = fn(); + ret.result = fn(); if (initcall_debug) { - it.rettime = ktime_get(); - delta = ktime_sub(it.rettime, it.calltime); - it.duration = (unsigned long long) delta.tv64 >> 10; + disable_boot_trace(); + rettime = ktime_get(); + delta = ktime_sub(rettime, calltime); + ret.duration = (unsigned long long) ktime_to_ns(delta) >> 10; + trace_boot_ret(&ret, fn); printk("initcall %pF returned %d after %Ld usecs\n", fn, - it.result, it.duration); - trace_boot(&it, fn); + ret.result, ret.duration); } msgbuf[0] = 0; - if (it.result && it.result != -ENODEV && initcall_debug) - sprintf(msgbuf, "error code %d ", it.result); + if (ret.result && ret.result != -ENODEV && initcall_debug) + sprintf(msgbuf, "error code %d ", ret.result); if (preempt_count() != count) { strlcat(msgbuf, "preemption imbalance ", sizeof(msgbuf)); @@ -741,7 +746,7 @@ int do_one_initcall(initcall_t fn) printk("initcall %pF returned with %s\n", fn, msgbuf); } - return it.result; + return ret.result; } @@ -882,7 +887,7 @@ static int __init kernel_init(void * unused) * we're essentially up and running. Get rid of the * initmem segments and start the user-mode stuff.. */ - stop_boot_trace(); + init_post(); return 0; } diff --git a/kernel/Makefile b/kernel/Makefile index 19fad003b19..010ccb31116 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -19,7 +19,10 @@ CFLAGS_REMOVE_mutex-debug.o = -pg CFLAGS_REMOVE_rtmutex-debug.o = -pg CFLAGS_REMOVE_cgroup-debug.o = -pg CFLAGS_REMOVE_sched_clock.o = -pg -CFLAGS_REMOVE_sched.o = -pg +endif +ifdef CONFIG_FUNCTION_RET_TRACER +CFLAGS_REMOVE_extable.o = -pg # For __kernel_text_address() +CFLAGS_REMOVE_module.o = -pg # For __module_text_address() endif obj-$(CONFIG_FREEZER) += freezer.o @@ -90,7 +93,7 @@ obj-$(CONFIG_FUNCTION_TRACER) += trace/ obj-$(CONFIG_TRACING) += trace/ obj-$(CONFIG_SMP) += sched_cpupri.o -ifneq ($(CONFIG_SCHED_NO_NO_OMIT_FRAME_POINTER),y) +ifneq ($(CONFIG_SCHED_OMIT_FRAME_POINTER),y) # According to Alan Modra <alan@linuxcare.com.au>, the -fno-omit-frame-pointer is # needed for x86 only. Why this used to be enabled for all architectures is beyond # me. I suspect most platforms don't need this, but until we know that for sure diff --git a/kernel/exit.c b/kernel/exit.c index 2d8be7ebb0f..61ba5b4b10c 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -53,6 +53,10 @@ #include <asm/pgtable.h> #include <asm/mmu_context.h> +DEFINE_TRACE(sched_process_free); +DEFINE_TRACE(sched_process_exit); +DEFINE_TRACE(sched_process_wait); + static void exit_mm(struct task_struct * tsk); static inline int task_detached(struct task_struct *p) @@ -1123,7 +1127,6 @@ NORET_TYPE void do_exit(long code) preempt_disable(); /* causes final put_task_struct in finish_task_switch(). */ tsk->state = TASK_DEAD; - schedule(); BUG(); /* Avoid "noreturn function does return". */ @@ -1321,10 +1324,10 @@ static int wait_task_zombie(struct task_struct *p, int options, * group, which consolidates times for all threads in the * group including the group leader. */ + thread_group_cputime(p, &cputime); spin_lock_irq(&p->parent->sighand->siglock); psig = p->parent->signal; sig = p->signal; - thread_group_cputime(p, &cputime); psig->cutime = cputime_add(psig->cutime, cputime_add(cputime.utime, diff --git a/kernel/extable.c b/kernel/extable.c index a26cb2e1702..adf0cc9c02d 100644 --- a/kernel/extable.c +++ b/kernel/extable.c @@ -66,3 +66,19 @@ int kernel_text_address(unsigned long addr) return 1; return module_text_address(addr) != NULL; } + +/* + * On some architectures (PPC64, IA64) function pointers + * are actually only tokens to some data that then holds the + * real function address. As a result, to find if a function + * pointer is part of the kernel text, we need to do some + * special dereferencing first. + */ +int func_ptr_is_kernel_text(void *ptr) +{ + unsigned long addr; + addr = (unsigned long) dereference_function_descriptor(ptr); + if (core_kernel_text(addr)) + return 1; + return module_text_address(addr) != NULL; +} diff --git a/kernel/fork.c b/kernel/fork.c index 2a372a0e206..d6e1a3205f6 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -47,6 +47,7 @@ #include <linux/mount.h> #include <linux/audit.h> #include <linux/memcontrol.h> +#include <linux/ftrace.h> #include <linux/profile.h> #include <linux/rmap.h> #include <linux/acct.h> @@ -80,6 +81,8 @@ DEFINE_PER_CPU(unsigned long, process_counts) = 0; __cacheline_aligned DEFINE_RWLOCK(tasklist_lock); /* outer */ +DEFINE_TRACE(sched_process_fork); + int nr_processes(void) { int cpu; @@ -137,6 +140,7 @@ void free_task(struct task_struct *tsk) prop_local_destroy_single(&tsk->dirties); free_thread_info(tsk->stack); rt_mutex_debug_task_free(tsk); + ftrace_retfunc_exit_task(tsk); free_task_struct(tsk); } EXPORT_SYMBOL(free_task); @@ -1267,6 +1271,7 @@ static struct task_struct *copy_process(unsigned long clone_flags, total_forks++; spin_unlock(¤t->sighand->siglock); write_unlock_irq(&tasklist_lock); + ftrace_retfunc_init_task(p); proc_fork_connector(p); cgroup_post_fork(p); return p; diff --git a/kernel/futex.c b/kernel/futex.c index 8af10027514..e10c5c8786a 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -123,24 +123,6 @@ struct futex_hash_bucket { static struct futex_hash_bucket futex_queues[1<<FUTEX_HASHBITS]; /* - * Take mm->mmap_sem, when futex is shared - */ -static inline void futex_lock_mm(struct rw_semaphore *fshared) -{ - if (fshared) - down_read(fshared); -} - -/* - * Release mm->mmap_sem, when the futex is shared - */ -static inline void futex_unlock_mm(struct rw_semaphore *fshared) -{ - if (fshared) - up_read(fshared); -} - -/* * We hash on the keys returned from get_futex_key (see below). */ static struct futex_hash_bucket *hash_futex(union futex_key *key) @@ -161,6 +143,45 @@ static inline int match_futex(union futex_key *key1, union futex_key *key2) && key1->both.offset == key2->both.offset); } +/* + * Take a reference to the resource addressed by a key. + * Can be called while holding spinlocks. + * + */ +static void get_futex_key_refs(union futex_key *key) +{ + if (!key->both.ptr) + return; + + switch (key->both.offset & (FUT_OFF_INODE|FUT_OFF_MMSHARED)) { + case FUT_OFF_INODE: + atomic_inc(&key->shared.inode->i_count); + break; + case FUT_OFF_MMSHARED: + atomic_inc(&key->private.mm->mm_count); + break; + } +} + +/* + * Drop a reference to the resource addressed by a key. + * The hash bucket spinlock must not be held. + */ +static void drop_futex_key_refs(union futex_key *key) +{ + if (!key->both.ptr) + return; + + switch (key->both.offset & (FUT_OFF_INODE|FUT_OFF_MMSHARED)) { + case FUT_OFF_INODE: + iput(key->shared.inode); + break; + case FUT_OFF_MMSHARED: + mmdrop(key->private.mm); + break; + } +} + /** * get_futex_key - Get parameters which are the keys for a futex. * @uaddr: virtual address of the futex @@ -179,12 +200,10 @@ static inline int match_futex(union futex_key *key1, union futex_key *key2) * For other futexes, it points to ¤t->mm->mmap_sem and * caller must have taken the reader lock. but NOT any spinlocks. */ -static int get_futex_key(u32 __user *uaddr, struct rw_semaphore *fshared, - union futex_key *key) +static int get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key) { unsigned long address = (unsigned long)uaddr; struct mm_struct *mm = current->mm; - struct vm_area_struct *vma; struct page *page; int err; @@ -208,100 +227,50 @@ static int get_futex_key(u32 __user *uaddr, struct rw_semaphore *fshared, return -EFAULT; key->private.mm = mm; key->private.address = address; + get_futex_key_refs(key); return 0; } - /* - * The futex is hashed differently depending on whether - * it's in a shared or private mapping. So check vma first. - */ - vma = find_extend_vma(mm, address); - if (unlikely(!vma)) - return -EFAULT; - /* - * Permissions. - */ - if (unlikely((vma->vm_flags & (VM_IO|VM_READ)) != VM_READ)) - return (vma->vm_flags & VM_IO) ? -EPERM : -EACCES; +again: + err = get_user_pages_fast(address, 1, 0, &page); + if (err < 0) + return err; + + lock_page(page); + if (!page->mapping) { + unlock_page(page); + put_page(page); + goto again; + } /* * Private mappings are handled in a simple way. * * NOTE: When userspace waits on a MAP_SHARED mapping, even if * it's a read-only handle, it's expected that futexes attach to - * the object not the particular process. Therefore we use - * VM_MAYSHARE here, not VM_SHARED which is restricted to shared - * mappings of _writable_ handles. + * the object not the particular process. */ - if (likely(!(vma->vm_flags & VM_MAYSHARE))) { - key->both.offset |= FUT_OFF_MMSHARED; /* reference taken on mm */ + if (PageAnon(page)) { + key->both.offset |= FUT_OFF_MMSHARED; /* ref taken on mm */ key->private.mm = mm; key->private.address = address; - return 0; + } else { + key->both.offset |= FUT_OFF_INODE; /* inode-based key */ + key->shared.inode = page->mapping->host; + key->shared.pgoff = page->index; } - /* - * Linear file mappings are also simple. - */ - key->shared.inode = vma->vm_file->f_path.dentry->d_inode; - key->both.offset |= FUT_OFF_INODE; /* inode-based key. */ - if (likely(!(vma->vm_flags & VM_NONLINEAR))) { - key->shared.pgoff = (((address - vma->vm_start) >> PAGE_SHIFT) - + vma->vm_pgoff); - return 0; - } + get_futex_key_refs(key); - /* - * We could walk the page table to read the non-linear - * pte, and get the page index without fetching the page - * from swap. But that's a lot of code to duplicate here - * for a rare case, so we simply fetch the page. - */ - err = get_user_pages(current, mm, address, 1, 0, 0, &page, NULL); - if (err >= 0) { - key->shared.pgoff = - page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT); - put_page(page); - return 0; - } - return err; -} - -/* - * Take a reference to the resource addressed by a key. - * Can be called while holding spinlocks. - * - */ -static void get_futex_key_refs(union futex_key *key) -{ - if (key->both.ptr == NULL) - return; - switch (key->both.offset & (FUT_OFF_INODE|FUT_OFF_MMSHARED)) { - case FUT_OFF_INODE: - atomic_inc(&key->shared.inode->i_count); - break; - case FUT_OFF_MMSHARED: - atomic_inc(&key->private.mm->mm_count); - break; - } + unlock_page(page); + put_page(page); + return 0; } -/* - * Drop a reference to the resource addressed by a key. - * The hash bucket spinlock must not be held. - */ -static void drop_futex_key_refs(union futex_key *key) +static inline +void put_futex_key(int fshared, union futex_key *key) { - if (!key->both.ptr) - return; - switch (key->both.offset & (FUT_OFF_INODE|FUT_OFF_MMSHARED)) { - case FUT_OFF_INODE: - iput(key->shared.inode); - break; - case FUT_OFF_MMSHARED: - mmdrop(key->private.mm); - break; - } + drop_futex_key_refs(key); } static u32 cmpxchg_futex_value_locked(u32 __user *uaddr, u32 uval, u32 newval) @@ -328,10 +297,8 @@ static int get_futex_value_locked(u32 *dest, u32 __user *from) /* * Fault handling. - * if fshared is non NULL, current->mm->mmap_sem is already held */ -static int futex_handle_fault(unsigned long address, - struct rw_semaphore *fshared, int attempt) +static int futex_handle_fault(unsigned long address, int attempt) { struct vm_area_struct * vma; struct mm_struct *mm = current->mm; @@ -340,8 +307,7 @@ static int futex_handle_fault(unsigned long address, if (attempt > 2) return ret; - if (!fshared) - down_read(&mm->mmap_sem); + down_read(&mm->mmap_sem); vma = find_vma(mm, address); if (vma && address >= vma->vm_start && (vma->vm_flags & VM_WRITE)) { @@ -361,8 +327,7 @@ static int futex_handle_fault(unsigned long address, current->min_flt++; } } - if (!fshared) - up_read(&mm->mmap_sem); + up_read(&mm->mmap_sem); return ret; } @@ -385,6 +350,7 @@ static int refill_pi_state_cache(void) /* pi_mutex gets initialized later */ pi_state->owner = NULL; atomic_set(&pi_state->refcount, 1); + pi_state->key = FUTEX_KEY_INIT; current->pi_state_cache = pi_state; @@ -462,7 +428,7 @@ void exit_pi_state_list(struct task_struct *curr) struct list_head *next, *head = &curr->pi_state_list; struct futex_pi_state *pi_state; struct futex_hash_bucket *hb; - union futex_key key; + union futex_key key = FUTEX_KEY_INIT; if (!futex_cmpxchg_enabled) return; @@ -719,20 +685,17 @@ double_lock_hb(struct futex_hash_bucket *hb1, struct futex_hash_bucket *hb2) * Wake up all waiters hashed on the physical page that is mapped * to this virtual address: */ -static int futex_wake(u32 __user *uaddr, struct rw_semaphore *fshared, - int nr_wake, u32 bitset) +static int futex_wake(u32 __user *uaddr, int fshared, int nr_wake, u32 bitset) { struct futex_hash_bucket *hb; struct futex_q *this, *next; struct plist_head *head; - union futex_key key; + union futex_key key = FUTEX_KEY_INIT; int ret; if (!bitset) return -EINVAL; - futex_lock_mm(fshared); - ret = get_futex_key(uaddr, fshared, &key); if (unlikely(ret != 0)) goto out; @@ -760,7 +723,7 @@ static int futex_wake(u32 __user *uaddr, struct rw_semaphore *fshared, spin_unlock(&hb->lock); out: - futex_unlock_mm(fshared); + put_futex_key(fshared, &key); return ret; } @@ -769,19 +732,16 @@ out: * to this virtual address: */ static int -futex_wake_op(u32 __user *uaddr1, struct rw_semaphore *fshared, - u32 __user *uaddr2, +futex_wake_op(u32 __user *uaddr1, int fshared, u32 __user *uaddr2, int nr_wake, int nr_wake2, int op) { - union futex_key key1, key2; + union futex_key key1 = FUTEX_KEY_INIT, key2 = FUTEX_KEY_INIT; struct futex_hash_bucket *hb1, *hb2; struct plist_head *head; struct futex_q *this, *next; int ret, op_ret, attempt = 0; retryfull: - futex_lock_mm(fshared); - ret = get_futex_key(uaddr1, fshared, &key1); if (unlikely(ret != 0)) goto out; @@ -826,18 +786,12 @@ retry: */ if (attempt++) { ret = futex_handle_fault((unsigned long)uaddr2, - fshared, attempt); + attempt); if (ret) goto out; goto retry; } - /* - * If we would have faulted, release mmap_sem, - * fault it in and start all over again. - */ - futex_unlock_mm(fshared); - ret = get_user(dummy, uaddr2); if (ret) return ret; @@ -873,7 +827,8 @@ retry: if (hb1 != hb2) spin_unlock(&hb2->lock); out: - futex_unlock_mm(fshared); + put_futex_key(fshared, &key2); + put_futex_key(fshared, &key1); return ret; } @@ -882,19 +837,16 @@ out: * Requeue all waiters hashed on one physical page to another * physical page. */ -static int futex_requeue(u32 __user *uaddr1, struct rw_semaphore *fshared, - u32 __user *uaddr2, +static int futex_requeue(u32 __user *uaddr1, int fshared, u32 __user *uaddr2, int nr_wake, int nr_requeue, u32 *cmpval) { - union futex_key key1, key2; + union futex_key key1 = FUTEX_KEY_INIT, key2 = FUTEX_KEY_INIT; struct futex_hash_bucket *hb1, *hb2; struct plist_head *head1; struct futex_q *this, *next; int ret, drop_count = 0; retry: - futex_lock_mm(fshared); - ret = get_futex_key(uaddr1, fshared, &key1); if (unlikely(ret != 0)) goto out; @@ -917,12 +869,6 @@ static int futex_requeue(u32 __user *uaddr1, struct rw_semaphore *fshared, if (hb1 != hb2) spin_unlock(&hb2->lock); - /* - * If we would have faulted, release mmap_sem, fault - * it in and start all over again. - */ - futex_unlock_mm(fshared); - ret = get_user(curval, uaddr1); if (!ret) @@ -974,7 +920,8 @@ out_unlock: drop_futex_key_refs(&key1); out: - futex_unlock_mm(fshared); + put_futex_key(fshared, &key2); + put_futex_key(fshared, &key1); return ret; } @@ -1096,8 +1043,7 @@ static void unqueue_me_pi(struct futex_q *q) * private futexes. */ static int fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q, - struct task_struct *newowner, - struct rw_semaphore *fshared) + struct task_struct *newowner, int fshared) { u32 newtid = task_pid_vnr(newowner) | FUTEX_WAITERS; struct futex_pi_state *pi_state = q->pi_state; @@ -1176,7 +1122,7 @@ retry: handle_fault: spin_unlock(q->lock_ptr); - ret = futex_handle_fault((unsigned long)uaddr, fshared, attempt++); + ret = futex_handle_fault((unsigned long)uaddr, attempt++); spin_lock(q->lock_ptr); @@ -1200,7 +1146,7 @@ handle_fault: static long futex_wait_restart(struct restart_block *restart); -static int futex_wait(u32 __user *uaddr, struct rw_semaphore *fshared, +static int futex_wait(u32 __user *uaddr, int fshared, u32 val, ktime_t *abs_time, u32 bitset) { struct task_struct *curr = current; @@ -1218,8 +1164,7 @@ static int futex_wait(u32 __user *uaddr, struct rw_semaphore *fshared, q.pi_state = NULL; q.bitset = bitset; retry: - futex_lock_mm(fshared); - + q.key = FUTEX_KEY_INIT; ret = get_futex_key(uaddr, fshared, &q.key); if (unlikely(ret != 0)) goto out_release_sem; @@ -1251,12 +1196,6 @@ static int futex_wait(u32 __user *uaddr, struct rw_semaphore *fshared, if (unlikely(ret)) { queue_unlock(&q, hb); - /* - * If we would have faulted, release mmap_sem, fault it in and - * start all over again. - */ - futex_unlock_mm(fshared); - ret = get_user(uval, uaddr); if (!ret) @@ -1271,12 +1210,6 @@ static int futex_wait(u32 __user *uaddr, struct rw_semaphore *fshared, queue_me(&q, hb); /* - * Now the futex is queued and we have checked the data, we - * don't want to hold mmap_sem while we sleep. - */ - futex_unlock_mm(fshared); - - /* * There might have been scheduling since the queue_me(), as we * cannot hold a spinlock across the get_user() in case it * faults, and we cannot just set TASK_INTERRUPTIBLE state when @@ -1363,7 +1296,7 @@ static int futex_wait(u32 __user *uaddr, struct rw_semaphore *fshared, queue_unlock(&q, hb); out_release_sem: - futex_unlock_mm(fshared); + put_futex_key(fshared, &q.key); return ret; } @@ -1371,13 +1304,13 @@ static int futex_wait(u32 __user *uaddr, struct rw_semaphore *fshared, static long futex_wait_restart(struct restart_block *restart) { u32 __user *uaddr = (u32 __user *)restart->futex.uaddr; - struct rw_semaphore *fshared = NULL; + int fshared = 0; ktime_t t; t.tv64 = restart->futex.time; restart->fn = do_no_restart_syscall; if (restart->futex.flags & FLAGS_SHARED) - fshared = ¤t->mm->mmap_sem; + fshared = 1; return (long)futex_wait(uaddr, fshared, restart->futex.val, &t, restart->futex.bitset); } @@ -1389,7 +1322,7 @@ static long futex_wait_restart(struct restart_block *restart) * if there are waiters then it will block, it does PI, etc. (Due to * races the kernel might see a 0 value of the futex too.) */ -static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, +static int futex_lock_pi(u32 __user *uaddr, int fshared, int detect, ktime_t *time, int trylock) { struct hrtimer_sleeper timeout, *to = NULL; @@ -1412,8 +1345,7 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, q.pi_state = NULL; retry: - futex_lock_mm(fshared); - + q.key = FUTEX_KEY_INIT; ret = get_futex_key(uaddr, fshared, &q.key); if (unlikely(ret != 0)) goto out_release_sem; @@ -1502,7 +1434,6 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, * exit to complete. */ queue_unlock(&q, hb); - futex_unlock_mm(fshared); cond_resched(); goto retry; @@ -1534,12 +1465,6 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, */ queue_me(&q, hb); - /* - * Now the futex is queued and we have checked the data, we - * don't want to hold mmap_sem while we sleep. - */ - futex_unlock_mm(fshared); - WARN_ON(!q.pi_state); /* * Block on the PI mutex: @@ -1552,7 +1477,6 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, ret = ret ? 0 : -EWOULDBLOCK; } - futex_lock_mm(fshared); spin_lock(q.lock_ptr); if (!ret) { @@ -1618,7 +1542,6 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, /* Unqueue and drop the lock */ unqueue_me_pi(&q); - futex_unlock_mm(fshared); if (to) destroy_hrtimer_on_stack(&to->timer); @@ -1628,7 +1551,7 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, queue_unlock(&q, hb); out_release_sem: - futex_unlock_mm(fshared); + put_futex_key(fshared, &q.key); if (to) destroy_hrtimer_on_stack(&to->timer); return ret; @@ -1645,15 +1568,12 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, queue_unlock(&q, hb); if (attempt++) { - ret = futex_handle_fault((unsigned long)uaddr, fshared, - attempt); + ret = futex_handle_fault((unsigned long)uaddr, attempt); if (ret) goto out_release_sem; goto retry_unlocked; } - futex_unlock_mm(fshared); - ret = get_user(uval, uaddr); if (!ret && (uval != -EFAULT)) goto retry; @@ -1668,13 +1588,13 @@ static int futex_lock_pi(u32 __user *uaddr, struct rw_semaphore *fshared, * This is the in-kernel slowpath: we look up the PI state (if any), * and do the rt-mutex unlock. */ -static int futex_unlock_pi(u32 __user *uaddr, struct rw_semaphore *fshared) +static int futex_unlock_pi(u32 __user *uaddr, int fshared) { struct futex_hash_bucket *hb; struct futex_q *this, *next; u32 uval; struct plist_head *head; - union futex_key key; + union futex_key key = FUTEX_KEY_INIT; int ret, attempt = 0; retry: @@ -1685,10 +1605,6 @@ retry: */ if ((uval & FUTEX_TID_MASK) != task_pid_vnr(current)) return -EPERM; - /* - * First take all the futex related locks: - */ - futex_lock_mm(fshared); ret = get_futex_key(uaddr, fshared, &key); if (unlikely(ret != 0)) @@ -1747,7 +1663,7 @@ retry_unlocked: out_unlock: spin_unlock(&hb->lock); out: - futex_unlock_mm(fshared); + put_futex_key(fshared, &key); return ret; @@ -1763,16 +1679,13 @@ pi_faulted: spin_unlock(&hb->lock); if (attempt++) { - ret = futex_handle_fault((unsigned long)uaddr, fshared, - attempt); + ret = futex_handle_fault((unsigned long)uaddr, attempt); if (ret) goto out; uval = 0; goto retry_unlocked; } - futex_unlock_mm(fshared); - ret = get_user(uval, uaddr); if (!ret && (uval != -EFAULT)) goto retry; @@ -1898,8 +1811,7 @@ retry: * PI futexes happens in exit_pi_state(): */ if (!pi && (uval & FUTEX_WAITERS)) - futex_wake(uaddr, &curr->mm->mmap_sem, 1, - FUTEX_BITSET_MATCH_ANY); + futex_wake(uaddr, 1, 1, FUTEX_BITSET_MATCH_ANY); } return 0; } @@ -1995,10 +1907,10 @@ long do_futex(u32 __user *uaddr, int op, u32 val, ktime_t *timeout, { int ret = -ENOSYS; int cmd = op & FUTEX_CMD_MASK; - struct rw_semaphore *fshared = NULL; + int fshared = 0; if (!(op & FUTEX_PRIVATE_FLAG)) - fshared = ¤t->mm->mmap_sem; + fshared = 1; switch (cmd) { case FUTEX_WAIT: diff --git a/kernel/kthread.c b/kernel/kthread.c index 8e7a7ce3ed0..4fbc456f393 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -21,6 +21,9 @@ static DEFINE_SPINLOCK(kthread_create_lock); static LIST_HEAD(kthread_create_list); struct task_struct *kthreadd_task; +DEFINE_TRACE(sched_kthread_stop); +DEFINE_TRACE(sched_kthread_stop_ret); + struct kthread_create_info { /* Information passed to kthread() from kthreadd. */ diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 46a404173db..e4bdda8dcf0 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -136,16 +136,16 @@ static inline struct lock_class *hlock_class(struct held_lock *hlock) #ifdef CONFIG_LOCK_STAT static DEFINE_PER_CPU(struct lock_class_stats[MAX_LOCKDEP_KEYS], lock_stats); -static int lock_contention_point(struct lock_class *class, unsigned long ip) +static int lock_point(unsigned long points[], unsigned long ip) { int i; - for (i = 0; i < ARRAY_SIZE(class->contention_point); i++) { - if (class->contention_point[i] == 0) { - class->contention_point[i] = ip; + for (i = 0; i < LOCKSTAT_POINTS; i++) { + if (points[i] == 0) { + points[i] = ip; break; } - if (class->contention_point[i] == ip) + if (points[i] == ip) break; } @@ -185,6 +185,9 @@ struct lock_class_stats lock_stats(struct lock_class *class) for (i = 0; i < ARRAY_SIZE(stats.contention_point); i++) stats.contention_point[i] += pcs->contention_point[i]; + for (i = 0; i < ARRAY_SIZE(stats.contending_point); i++) + stats.contending_point[i] += pcs->contending_point[i]; + lock_time_add(&pcs->read_waittime, &stats.read_waittime); lock_time_add(&pcs->write_waittime, &stats.write_waittime); @@ -209,6 +212,7 @@ void clear_lock_stats(struct lock_class *class) memset(cpu_stats, 0, sizeof(struct lock_class_stats)); } memset(class->contention_point, 0, sizeof(class->contention_point)); + memset(class->contending_point, 0, sizeof(class->contending_point)); } static struct lock_class_stats *get_lock_stats(struct lock_class *class) @@ -2999,7 +3003,7 @@ __lock_contended(struct lockdep_map *lock, unsigned long ip) struct held_lock *hlock, *prev_hlock; struct lock_class_stats *stats; unsigned int depth; - int i, point; + int i, contention_point, contending_point; depth = curr->lockdep_depth; if (DEBUG_LOCKS_WARN_ON(!depth)) @@ -3023,18 +3027,22 @@ __lock_contended(struct lockdep_map *lock, unsigned long ip) found_it: hlock->waittime_stamp = sched_clock(); - point = lock_contention_point(hlock_class(hlock), ip); + contention_point = lock_point(hlock_class(hlock)->contention_point, ip); + contending_point = lock_point(hlock_class(hlock)->contending_point, + lock->ip); stats = get_lock_stats(hlock_class(hlock)); - if (point < ARRAY_SIZE(stats->contention_point)) - stats->contention_point[point]++; + if (contention_point < LOCKSTAT_POINTS) + stats->contention_point[contention_point]++; + if (contending_point < LOCKSTAT_POINTS) + stats->contending_point[contending_point]++; if (lock->cpu != smp_processor_id()) stats->bounces[bounce_contended + !!hlock->read]++; put_lock_stats(stats); } static void -__lock_acquired(struct lockdep_map *lock) +__lock_acquired(struct lockdep_map *lock, unsigned long ip) { struct task_struct *curr = current; struct held_lock *hlock, *prev_hlock; @@ -3083,6 +3091,7 @@ found_it: put_lock_stats(stats); lock->cpu = cpu; + lock->ip = ip; } void lock_contended(struct lockdep_map *lock, unsigned long ip) @@ -3104,7 +3113,7 @@ void lock_contended(struct lockdep_map *lock, unsigned long ip) } EXPORT_SYMBOL_GPL(lock_contended); -void lock_acquired(struct lockdep_map *lock) +void lock_acquired(struct lockdep_map *lock, unsigned long ip) { unsigned long flags; @@ -3117,7 +3126,7 @@ void lock_acquired(struct lockdep_map *lock) raw_local_irq_save(flags); check_flags(flags); current->lockdep_recursion = 1; - __lock_acquired(lock); + __lock_acquired(lock, ip); current->lockdep_recursion = 0; raw_local_irq_restore(flags); } diff --git a/kernel/lockdep_proc.c b/kernel/lockdep_proc.c index 20dbcbf9c7d..13716b81389 100644 --- a/kernel/lockdep_proc.c +++ b/kernel/lockdep_proc.c @@ -470,11 +470,12 @@ static void seq_line(struct seq_file *m, char c, int offset, int length) static void snprint_time(char *buf, size_t bufsiz, s64 nr) { - unsigned long rem; + s64 div; + s32 rem; nr += 5; /* for display rounding */ - rem = do_div(nr, 1000); /* XXX: do_div_signed */ - snprintf(buf, bufsiz, "%lld.%02d", (long long)nr, (int)rem/10); + div = div_s64_rem(nr, 1000, &rem); + snprintf(buf, bufsiz, "%lld.%02d", (long long)div, (int)rem/10); } static void seq_time(struct seq_file *m, s64 time) @@ -556,7 +557,7 @@ static void seq_stats(struct seq_file *m, struct lock_stat_data *data) if (stats->read_holdtime.nr) namelen += 2; - for (i = 0; i < ARRAY_SIZE(class->contention_point); i++) { + for (i = 0; i < LOCKSTAT_POINTS; i++) { char sym[KSYM_SYMBOL_LEN]; char ip[32]; @@ -573,6 +574,23 @@ static void seq_stats(struct seq_file *m, struct lock_stat_data *data) stats->contention_point[i], ip, sym); } + for (i = 0; i < LOCKSTAT_POINTS; i++) { + char sym[KSYM_SYMBOL_LEN]; + char ip[32]; + + if (class->contending_point[i] == 0) + break; + + if (!i) + seq_line(m, '-', 40-namelen, namelen); + + sprint_symbol(sym, class->contending_point[i]); + snprintf(ip, sizeof(ip), "[<%p>]", + (void *)class->contending_point[i]); + seq_printf(m, "%40s %14lu %29s %s\n", name, + stats->contending_point[i], + ip, sym); + } if (i) { seq_puts(m, "\n"); seq_line(m, '.', 0, 40 + 1 + 10 * (14 + 1)); @@ -582,7 +600,7 @@ static void seq_stats(struct seq_file *m, struct lock_stat_data *data) static void seq_header(struct seq_file *m) { - seq_printf(m, "lock_stat version 0.2\n"); + seq_printf(m, "lock_stat version 0.3\n"); seq_line(m, '-', 0, 40 + 1 + 10 * (14 + 1)); seq_printf(m, "%40s %14s %14s %14s %14s %14s %14s %14s %14s " "%14s %14s\n", diff --git a/kernel/marker.c b/kernel/marker.c index e9c6b2bc940..ea54f264786 100644 --- a/kernel/marker.c +++ b/kernel/marker.c @@ -43,6 +43,7 @@ static DEFINE_MUTEX(markers_mutex); */ #define MARKER_HASH_BITS 6 #define MARKER_TABLE_SIZE (1 << MARKER_HASH_BITS) +static struct hlist_head marker_table[MARKER_TABLE_SIZE]; /* * Note about RCU : @@ -64,11 +65,10 @@ struct marker_entry { void *oldptr; int rcu_pending; unsigned char ptype:1; + unsigned char format_allocated:1; char name[0]; /* Contains name'\0'format'\0' */ }; -static struct hlist_head marker_table[MARKER_TABLE_SIZE]; - /** * __mark_empty_function - Empty probe callback * @probe_private: probe private data @@ -81,7 +81,7 @@ static struct hlist_head marker_table[MARKER_TABLE_SIZE]; * though the function pointer change and the marker enabling are two distinct * operations that modifies the execution flow of preemptible code. */ -void __mark_empty_function(void *probe_private, void *call_private, +notrace void __mark_empty_function(void *probe_private, void *call_private, const char *fmt, va_list *args) { } @@ -97,7 +97,8 @@ EXPORT_SYMBOL_GPL(__mark_empty_function); * need to put a full smp_rmb() in this branch. This is why we do not use * rcu_dereference() for the pointer read. */ -void marker_probe_cb(const struct marker *mdata, void *call_private, ...) +notrace void marker_probe_cb(const struct marker *mdata, + void *call_private, ...) { va_list args; char ptype; @@ -107,7 +108,7 @@ void marker_probe_cb(const struct marker *mdata, void *call_private, ...) * sure the teardown of the callbacks can be done correctly when they * are in modules and they insure RCU read coherency. */ - rcu_read_lock_sched(); + rcu_read_lock_sched_notrace(); ptype = mdata->ptype; if (likely(!ptype)) { marker_probe_func *func; @@ -145,7 +146,7 @@ void marker_probe_cb(const struct marker *mdata, void *call_private, ...) va_end(args); } } - rcu_read_unlock_sched(); + rcu_read_unlock_sched_notrace(); } EXPORT_SYMBOL_GPL(marker_probe_cb); @@ -157,12 +158,13 @@ EXPORT_SYMBOL_GPL(marker_probe_cb); * * Should be connected to markers "MARK_NOARGS". */ -void marker_probe_cb_noarg(const struct marker *mdata, void *call_private, ...) +static notrace void marker_probe_cb_noarg(const struct marker *mdata, + void *call_private, ...) { va_list args; /* not initialized */ char ptype; - rcu_read_lock_sched(); + rcu_read_lock_sched_notrace(); ptype = mdata->ptype; if (likely(!ptype)) { marker_probe_func *func; @@ -195,9 +197,8 @@ void marker_probe_cb_noarg(const struct marker *mdata, void *call_private, ...) multi[i].func(multi[i].probe_private, call_private, mdata->format, &args); } - rcu_read_unlock_sched(); + rcu_read_unlock_sched_notrace(); } -EXPORT_SYMBOL_GPL(marker_probe_cb_noarg); static void free_old_closure(struct rcu_head *head) { @@ -416,6 +417,7 @@ static struct marker_entry *add_marker(const char *name, const char *format) e->single.probe_private = NULL; e->multi = NULL; e->ptype = 0; + e->format_allocated = 0; e->refcount = 0; e->rcu_pending = 0; hlist_add_head(&e->hlist, head); @@ -447,6 +449,8 @@ static int remove_marker(const char *name) if (e->single.func != __mark_empty_function) return -EBUSY; hlist_del(&e->hlist); + if (e->format_allocated) + kfree(e->format); /* Make sure the call_rcu has been executed */ if (e->rcu_pending) rcu_barrier_sched(); @@ -457,57 +461,34 @@ static int remove_marker(const char *name) /* * Set the mark_entry format to the format found in the element. */ -static int marker_set_format(struct marker_entry **entry, const char *format) +static int marker_set_format(struct marker_entry *entry, const char *format) { - struct marker_entry *e; - size_t name_len = strlen((*entry)->name) + 1; - size_t format_len = strlen(format) + 1; - - - e = kmalloc(sizeof(struct marker_entry) + name_len + format_len, - GFP_KERNEL); - if (!e) + entry->format = kstrdup(format, GFP_KERNEL); + if (!entry->format) return -ENOMEM; - memcpy(&e->name[0], (*entry)->name, name_len); - e->format = &e->name[name_len]; - memcpy(e->format, format, format_len); - if (strcmp(e->format, MARK_NOARGS) == 0) - e->call = marker_probe_cb_noarg; - else - e->call = marker_probe_cb; - e->single = (*entry)->single; - e->multi = (*entry)->multi; - e->ptype = (*entry)->ptype; - e->refcount = (*entry)->refcount; - e->rcu_pending = 0; - hlist_add_before(&e->hlist, &(*entry)->hlist); - hlist_del(&(*entry)->hlist); - /* Make sure the call_rcu has been executed */ - if ((*entry)->rcu_pending) - rcu_barrier_sched(); - kfree(*entry); - *entry = e; + entry->format_allocated = 1; + trace_mark(core_marker_format, "name %s format %s", - e->name, e->format); + entry->name, entry->format); return 0; } /* * Sets the probe callback corresponding to one marker. */ -static int set_marker(struct marker_entry **entry, struct marker *elem, +static int set_marker(struct marker_entry *entry, struct marker *elem, int active) { - int ret; - WARN_ON(strcmp((*entry)->name, elem->name) != 0); + int ret = 0; + WARN_ON(strcmp(entry->name, elem->name) != 0); - if ((*entry)->format) { - if (strcmp((*entry)->format, elem->format) != 0) { + if (entry->format) { + if (strcmp(entry->format, elem->format) != 0) { printk(KERN_NOTICE "Format mismatch for probe %s " "(%s), marker (%s)\n", - (*entry)->name, - (*entry)->format, + entry->name, + entry->format, elem->format); return -EPERM; } @@ -523,37 +504,67 @@ static int set_marker(struct marker_entry **entry, struct marker *elem, * pass from a "safe" callback (with argument) to an "unsafe" * callback (does not set arguments). */ - elem->call = (*entry)->call; + elem->call = entry->call; /* * Sanity check : * We only update the single probe private data when the ptr is * set to a _non_ single probe! (0 -> 1 and N -> 1, N != 1) */ WARN_ON(elem->single.func != __mark_empty_function - && elem->single.probe_private - != (*entry)->single.probe_private && - !elem->ptype); - elem->single.probe_private = (*entry)->single.probe_private; + && elem->single.probe_private != entry->single.probe_private + && !elem->ptype); + elem->single.probe_private = entry->single.probe_private; /* * Make sure the private data is valid when we update the * single probe ptr. */ smp_wmb(); - elem->single.func = (*entry)->single.func; + elem->single.func = entry->single.func; /* * We also make sure that the new probe callbacks array is consistent * before setting a pointer to it. */ - rcu_assign_pointer(elem->multi, (*entry)->multi); + rcu_assign_pointer(elem->multi, entry->multi); /* * Update the function or multi probe array pointer before setting the * ptype. */ smp_wmb(); - elem->ptype = (*entry)->ptype; + elem->ptype = entry->ptype; + + if (elem->tp_name && (active ^ elem->state)) { + WARN_ON(!elem->tp_cb); + /* + * It is ok to directly call the probe registration because type + * checking has been done in the __trace_mark_tp() macro. + */ + + if (active) { + /* + * try_module_get should always succeed because we hold + * lock_module() to get the tp_cb address. + */ + ret = try_module_get(__module_text_address( + (unsigned long)elem->tp_cb)); + BUG_ON(!ret); + ret = tracepoint_probe_register_noupdate( + elem->tp_name, + elem->tp_cb); + } else { + ret = tracepoint_probe_unregister_noupdate( + elem->tp_name, + elem->tp_cb); + /* + * tracepoint_probe_update_all() must be called + * before the module containing tp_cb is unloaded. + */ + module_put(__module_text_address( + (unsigned long)elem->tp_cb)); + } + } elem->state = active; - return 0; + return ret; } /* @@ -564,7 +575,24 @@ static int set_marker(struct marker_entry **entry, struct marker *elem, */ static void disable_marker(struct marker *elem) { + int ret; + /* leave "call" as is. It is known statically. */ + if (elem->tp_name && elem->state) { + WARN_ON(!elem->tp_cb); + /* + * It is ok to directly call the probe registration because type + * checking has been done in the __trace_mark_tp() macro. + */ + ret = tracepoint_probe_unregister_noupdate(elem->tp_name, + elem->tp_cb); + WARN_ON(ret); + /* + * tracepoint_probe_update_all() must be called + * before the module containing tp_cb is unloaded. + */ + module_put(__module_text_address((unsigned long)elem->tp_cb)); + } elem->state = 0; elem->single.func = __mark_empty_function; /* Update the function before setting the ptype */ @@ -594,8 +622,7 @@ void marker_update_probe_range(struct marker *begin, for (iter = begin; iter < end; iter++) { mark_entry = get_marker(iter->name); if (mark_entry) { - set_marker(&mark_entry, iter, - !!mark_entry->refcount); + set_marker(mark_entry, iter, !!mark_entry->refcount); /* * ignore error, continue */ @@ -629,6 +656,7 @@ static void marker_update_probes(void) marker_update_probe_range(__start___markers, __stop___markers); /* Markers in modules. */ module_update_markers(); + tracepoint_probe_update_all(); } /** @@ -657,7 +685,7 @@ int marker_probe_register(const char *name, const char *format, ret = PTR_ERR(entry); } else if (format) { if (!entry->format) - ret = marker_set_format(&entry, format); + ret = marker_set_format(entry, format); else if (strcmp(entry->format, format)) ret = -EPERM; } @@ -676,10 +704,11 @@ int marker_probe_register(const char *name, const char *format, goto end; } mutex_unlock(&markers_mutex); - marker_update_probes(); /* may update entry */ + marker_update_probes(); mutex_lock(&markers_mutex); entry = get_marker(name); - WARN_ON(!entry); + if (!entry) + goto end; if (entry->rcu_pending) rcu_barrier_sched(); entry->oldptr = old; @@ -720,7 +749,7 @@ int marker_probe_unregister(const char *name, rcu_barrier_sched(); old = marker_entry_remove_probe(entry, probe, probe_private); mutex_unlock(&markers_mutex); - marker_update_probes(); /* may update entry */ + marker_update_probes(); mutex_lock(&markers_mutex); entry = get_marker(name); if (!entry) @@ -801,10 +830,11 @@ int marker_probe_unregister_private_data(marker_probe_func *probe, rcu_barrier_sched(); old = marker_entry_remove_probe(entry, NULL, probe_private); mutex_unlock(&markers_mutex); - marker_update_probes(); /* may update entry */ + marker_update_probes(); mutex_lock(&markers_mutex); entry = get_marker_from_private_data(probe, probe_private); - WARN_ON(!entry); + if (!entry) + goto end; if (entry->rcu_pending) rcu_barrier_sched(); entry->oldptr = old; @@ -848,8 +878,6 @@ void *marker_get_private_data(const char *name, marker_probe_func *probe, if (!e->ptype) { if (num == 0 && e->single.func == probe) return e->single.probe_private; - else - break; } else { struct marker_probe_closure *closure; int match = 0; @@ -861,8 +889,42 @@ void *marker_get_private_data(const char *name, marker_probe_func *probe, return closure[i].probe_private; } } + break; } } return ERR_PTR(-ENOENT); } EXPORT_SYMBOL_GPL(marker_get_private_data); + +#ifdef CONFIG_MODULES + +int marker_module_notify(struct notifier_block *self, + unsigned long val, void *data) +{ + struct module *mod = data; + + switch (val) { + case MODULE_STATE_COMING: + marker_update_probe_range(mod->markers, + mod->markers + mod->num_markers); + break; + case MODULE_STATE_GOING: + marker_update_probe_range(mod->markers, + mod->markers + mod->num_markers); + break; + } + return 0; +} + +struct notifier_block marker_module_nb = { + .notifier_call = marker_module_notify, + .priority = 0, +}; + +static int init_markers(void) +{ + return register_module_notifier(&marker_module_nb); +} +__initcall(init_markers); + +#endif /* CONFIG_MODULES */ diff --git a/kernel/module.c b/kernel/module.c index 1f4cc00e0c2..89bcf7c1327 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -2184,24 +2184,15 @@ static noinline struct module *load_module(void __user *umod, struct mod_debug *debug; unsigned int num_debug; -#ifdef CONFIG_MARKERS - marker_update_probe_range(mod->markers, - mod->markers + mod->num_markers); -#endif debug = section_objs(hdr, sechdrs, secstrings, "__verbose", sizeof(*debug), &num_debug); dynamic_printk_setup(debug, num_debug); - -#ifdef CONFIG_TRACEPOINTS - tracepoint_update_probe_range(mod->tracepoints, - mod->tracepoints + mod->num_tracepoints); -#endif } /* sechdrs[0].sh_size is always zero */ mseg = section_objs(hdr, sechdrs, secstrings, "__mcount_loc", sizeof(*mseg), &num_mcount); - ftrace_init_module(mseg, mseg + num_mcount); + ftrace_init_module(mod, mseg, mseg + num_mcount); err = module_finalize(hdr, sechdrs, mod); if (err < 0) diff --git a/kernel/mutex.c b/kernel/mutex.c index 12c779dc65d..4f45d4b658e 100644 --- a/kernel/mutex.c +++ b/kernel/mutex.c @@ -59,7 +59,7 @@ EXPORT_SYMBOL(__mutex_init); * We also put the fastpath first in the kernel image, to make sure the * branch is predicted by the CPU as default-untaken. */ -static void noinline __sched +static __used noinline void __sched __mutex_lock_slowpath(atomic_t *lock_count); /*** @@ -96,7 +96,7 @@ void inline __sched mutex_lock(struct mutex *lock) EXPORT_SYMBOL(mutex_lock); #endif -static noinline void __sched __mutex_unlock_slowpath(atomic_t *lock_count); +static __used noinline void __sched __mutex_unlock_slowpath(atomic_t *lock_count); /*** * mutex_unlock - release the mutex @@ -184,7 +184,7 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass, } done: - lock_acquired(&lock->dep_map); + lock_acquired(&lock->dep_map, ip); /* got the lock - rejoice! */ mutex_remove_waiter(lock, &waiter, task_thread_info(task)); debug_mutex_set_owner(lock, task_thread_info(task)); @@ -268,7 +268,7 @@ __mutex_unlock_common_slowpath(atomic_t *lock_count, int nested) /* * Release the lock, slowpath: */ -static noinline void +static __used noinline void __mutex_unlock_slowpath(atomic_t *lock_count) { __mutex_unlock_common_slowpath(lock_count, 1); @@ -313,7 +313,7 @@ int __sched mutex_lock_killable(struct mutex *lock) } EXPORT_SYMBOL(mutex_lock_killable); -static noinline void __sched +static __used noinline void __sched __mutex_lock_slowpath(atomic_t *lock_count) { struct mutex *lock = container_of(lock_count, struct mutex, count); diff --git a/kernel/notifier.c b/kernel/notifier.c index 4282c0a40a5..61d5aa5eced 100644 --- a/kernel/notifier.c +++ b/kernel/notifier.c @@ -82,6 +82,14 @@ static int __kprobes notifier_call_chain(struct notifier_block **nl, while (nb && nr_to_call) { next_nb = rcu_dereference(nb->next); + +#ifdef CONFIG_DEBUG_NOTIFIERS + if (unlikely(!func_ptr_is_kernel_text(nb->notifier_call))) { + WARN(1, "Invalid notifier called!"); + nb = next_nb; + continue; + } +#endif ret = nb->notifier_call(nb, val, v); if (nr_calls) diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 895337b16a2..3f4377e0aa0 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c @@ -58,21 +58,21 @@ void thread_group_cputime( struct task_struct *tsk, struct task_cputime *times) { - struct signal_struct *sig; + struct task_cputime *totals, *tot; int i; - struct task_cputime *tot; - sig = tsk->signal; - if (unlikely(!sig) || !sig->cputime.totals) { + totals = tsk->signal->cputime.totals; + if (!totals) { times->utime = tsk->utime; times->stime = tsk->stime; times->sum_exec_runtime = tsk->se.sum_exec_runtime; return; } + times->stime = times->utime = cputime_zero; times->sum_exec_runtime = 0; for_each_possible_cpu(i) { - tot = per_cpu_ptr(tsk->signal->cputime.totals, i); + tot = per_cpu_ptr(totals, i); times->utime = cputime_add(times->utime, tot->utime); times->stime = cputime_add(times->stime, tot->stime); times->sum_exec_runtime += tot->sum_exec_runtime; diff --git a/kernel/power/disk.c b/kernel/power/disk.c index c9d74083746..f77d3819ef5 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -22,7 +22,6 @@ #include <linux/console.h> #include <linux/cpu.h> #include <linux/freezer.h> -#include <linux/ftrace.h> #include "power.h" @@ -257,7 +256,7 @@ static int create_image(int platform_mode) int hibernation_snapshot(int platform_mode) { - int error, ftrace_save; + int error; /* Free memory before shutting down devices. */ error = swsusp_shrink_memory(); @@ -269,7 +268,6 @@ int hibernation_snapshot(int platform_mode) goto Close; suspend_console(); - ftrace_save = __ftrace_enabled_save(); error = device_suspend(PMSG_FREEZE); if (error) goto Recover_platform; @@ -299,7 +297,6 @@ int hibernation_snapshot(int platform_mode) Resume_devices: device_resume(in_suspend ? (error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE); - __ftrace_enabled_restore(ftrace_save); resume_console(); Close: platform_end(platform_mode); @@ -370,11 +367,10 @@ static int resume_target_kernel(void) int hibernation_restore(int platform_mode) { - int error, ftrace_save; + int error; pm_prepare_console(); suspend_console(); - ftrace_save = __ftrace_enabled_save(); error = device_suspend(PMSG_QUIESCE); if (error) goto Finish; @@ -389,7 +385,6 @@ int hibernation_restore(int platform_mode) platform_restore_cleanup(platform_mode); device_resume(PMSG_RECOVER); Finish: - __ftrace_enabled_restore(ftrace_save); resume_console(); pm_restore_console(); return error; @@ -402,7 +397,7 @@ int hibernation_restore(int platform_mode) int hibernation_platform_enter(void) { - int error, ftrace_save; + int error; if (!hibernation_ops) return -ENOSYS; @@ -417,7 +412,6 @@ int hibernation_platform_enter(void) goto Close; suspend_console(); - ftrace_save = __ftrace_enabled_save(); error = device_suspend(PMSG_HIBERNATE); if (error) { if (hibernation_ops->recover) @@ -452,7 +446,6 @@ int hibernation_platform_enter(void) hibernation_ops->finish(); Resume_devices: device_resume(PMSG_RESTORE); - __ftrace_enabled_restore(ftrace_save); resume_console(); Close: hibernation_ops->end(); diff --git a/kernel/power/main.c b/kernel/power/main.c index b8f7ce9473e..613f16941b8 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -22,7 +22,6 @@ #include <linux/freezer.h> #include <linux/vmstat.h> #include <linux/syscalls.h> -#include <linux/ftrace.h> #include "power.h" @@ -317,7 +316,7 @@ static int suspend_enter(suspend_state_t state) */ int suspend_devices_and_enter(suspend_state_t state) { - int error, ftrace_save; + int error; if (!suspend_ops) return -ENOSYS; @@ -328,7 +327,6 @@ int suspend_devices_and_enter(suspend_state_t state) goto Close; } suspend_console(); - ftrace_save = __ftrace_enabled_save(); suspend_test_start(); error = device_suspend(PMSG_SUSPEND); if (error) { @@ -360,7 +358,6 @@ int suspend_devices_and_enter(suspend_state_t state) suspend_test_start(); device_resume(PMSG_RESUME); suspend_test_finish("resume devices"); - __ftrace_enabled_restore(ftrace_save); resume_console(); Close: if (suspend_ops->end) diff --git a/kernel/profile.c b/kernel/profile.c index dc41827fbfe..60adefb59b5 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -544,7 +544,7 @@ static const struct file_operations proc_profile_operations = { }; #ifdef CONFIG_SMP -static inline void profile_nop(void *unused) +static void profile_nop(void *unused) { } diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c index 37f72e55154..c03ca3e6191 100644 --- a/kernel/rcuclassic.c +++ b/kernel/rcuclassic.c @@ -191,7 +191,7 @@ static void print_other_cpu_stall(struct rcu_ctrlblk *rcp) /* OK, time to rat on our buddy... */ - printk(KERN_ERR "RCU detected CPU stalls:"); + printk(KERN_ERR "INFO: RCU detected CPU stalls:"); for_each_possible_cpu(cpu) { if (cpu_isset(cpu, rcp->cpumask)) printk(" %d", cpu); @@ -204,7 +204,7 @@ static void print_cpu_stall(struct rcu_ctrlblk *rcp) { unsigned long flags; - printk(KERN_ERR "RCU detected CPU %d stall (t=%lu/%lu jiffies)\n", + printk(KERN_ERR "INFO: RCU detected CPU %d stall (t=%lu/%lu jiffies)\n", smp_processor_id(), jiffies, jiffies - rcp->gp_start); dump_stack(); @@ -393,7 +393,7 @@ static void rcu_start_batch(struct rcu_ctrlblk *rcp) * unnecessarily. */ smp_mb(); - cpus_andnot(rcp->cpumask, cpu_online_map, nohz_cpu_mask); + cpumask_andnot(&rcp->cpumask, cpu_online_mask, nohz_cpu_mask); rcp->signaled = 0; } diff --git a/kernel/sched.c b/kernel/sched.c index b7480fb5c3d..8050a61a7ad 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -118,6 +118,12 @@ */ #define RUNTIME_INF ((u64)~0ULL) +DEFINE_TRACE(sched_wait_task); +DEFINE_TRACE(sched_wakeup); +DEFINE_TRACE(sched_wakeup_new); +DEFINE_TRACE(sched_switch); +DEFINE_TRACE(sched_migrate_task); + #ifdef CONFIG_SMP /* * Divide a load by a sched group cpu_power : (load / sg->__cpu_power) @@ -481,14 +487,14 @@ struct rt_rq { */ struct root_domain { atomic_t refcount; - cpumask_t span; - cpumask_t online; + cpumask_var_t span; + cpumask_var_t online; /* * The "RT overload" flag: it gets set if a CPU has more than * one runnable RT task. */ - cpumask_t rto_mask; + cpumask_var_t rto_mask; atomic_t rto_count; #ifdef CONFIG_SMP struct cpupri cpupri; @@ -703,45 +709,18 @@ static __read_mostly char *sched_feat_names[] = { #undef SCHED_FEAT -static int sched_feat_open(struct inode *inode, struct file *filp) +static int sched_feat_show(struct seq_file *m, void *v) { - filp->private_data = inode->i_private; - return 0; -} - -static ssize_t -sched_feat_read(struct file *filp, char __user *ubuf, - size_t cnt, loff_t *ppos) -{ - char *buf; - int r = 0; - int len = 0; int i; for (i = 0; sched_feat_names[i]; i++) { - len += strlen(sched_feat_names[i]); - len += 4; + if (!(sysctl_sched_features & (1UL << i))) + seq_puts(m, "NO_"); + seq_printf(m, "%s ", sched_feat_names[i]); } + seq_puts(m, "\n"); - buf = kmalloc(len + 2, GFP_KERNEL); - if (!buf) - return -ENOMEM; - - for (i = 0; sched_feat_names[i]; i++) { - if (sysctl_sched_features & (1UL << i)) - r += sprintf(buf + r, "%s ", sched_feat_names[i]); - else - r += sprintf(buf + r, "NO_%s ", sched_feat_names[i]); - } - - r += sprintf(buf + r, "\n"); - WARN_ON(r >= len + 2); - - r = simple_read_from_buffer(ubuf, cnt, ppos, buf, r); - - kfree(buf); - - return r; + return 0; } static ssize_t @@ -786,10 +765,17 @@ sched_feat_write(struct file *filp, const char __user *ubuf, return cnt; } +static int sched_feat_open(struct inode *inode, struct file *filp) +{ + return single_open(filp, sched_feat_show, NULL); +} + static struct file_operations sched_feat_fops = { - .open = sched_feat_open, - .read = sched_feat_read, - .write = sched_feat_write, + .open = sched_feat_open, + .write = sched_feat_write, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, }; static __init int sched_init_debug(void) @@ -1474,27 +1460,13 @@ static void update_group_shares_cpu(struct task_group *tg, int cpu, unsigned long sd_shares, unsigned long sd_rq_weight) { - int boost = 0; unsigned long shares; unsigned long rq_weight; if (!tg->se[cpu]) return; - rq_weight = tg->cfs_rq[cpu]->load.weight; - - /* - * If there are currently no tasks on the cpu pretend there is one of - * average load so that when a new task gets to run here it will not - * get delayed by group starvation. - */ - if (!rq_weight) { - boost = 1; - rq_weight = NICE_0_LOAD; - } - - if (unlikely(rq_weight > sd_rq_weight)) - rq_weight = sd_rq_weight; + rq_weight = tg->cfs_rq[cpu]->rq_weight; /* * \Sum shares * rq_weight @@ -1502,7 +1474,7 @@ update_group_shares_cpu(struct task_group *tg, int cpu, * \Sum rq_weight * */ - shares = (sd_shares * rq_weight) / (sd_rq_weight + 1); + shares = (sd_shares * rq_weight) / sd_rq_weight; shares = clamp_t(unsigned long, shares, MIN_SHARES, MAX_SHARES); if (abs(shares - tg->se[cpu]->load.weight) > @@ -1511,11 +1483,7 @@ update_group_shares_cpu(struct task_group *tg, int cpu, unsigned long flags; spin_lock_irqsave(&rq->lock, flags); - /* - * record the actual number of shares, not the boosted amount. - */ - tg->cfs_rq[cpu]->shares = boost ? 0 : shares; - tg->cfs_rq[cpu]->rq_weight = rq_weight; + tg->cfs_rq[cpu]->shares = shares; __set_se_shares(tg->se[cpu], shares); spin_unlock_irqrestore(&rq->lock, flags); @@ -1529,13 +1497,23 @@ update_group_shares_cpu(struct task_group *tg, int cpu, */ static int tg_shares_up(struct task_group *tg, void *data) { - unsigned long rq_weight = 0; + unsigned long weight, rq_weight = 0; unsigned long shares = 0; struct sched_domain *sd = data; int i; - for_each_cpu_mask(i, sd->span) { - rq_weight += tg->cfs_rq[i]->load.weight; + for_each_cpu(i, sched_domain_span(sd)) { + /* + * If there are currently no tasks on the cpu pretend there + * is one of average load so that when a new task gets to + * run here it will not get delayed by group starvation. + */ + weight = tg->cfs_rq[i]->load.weight; + if (!weight) + weight = NICE_0_LOAD; + + tg->cfs_rq[i]->rq_weight = weight; + rq_weight += weight; shares += tg->cfs_rq[i]->shares; } @@ -1545,10 +1523,7 @@ static int tg_shares_up(struct task_group *tg, void *data) if (!sd->parent || !(sd->parent->flags & SD_LOAD_BALANCE)) shares = tg->shares; - if (!rq_weight) - rq_weight = cpus_weight(sd->span) * NICE_0_LOAD; - - for_each_cpu_mask(i, sd->span) + for_each_cpu(i, sched_domain_span(sd)) update_group_shares_cpu(tg, i, shares, rq_weight); return 0; @@ -2079,15 +2054,17 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu) int i; /* Skip over this group if it has no CPUs allowed */ - if (!cpus_intersects(group->cpumask, p->cpus_allowed)) + if (!cpumask_intersects(sched_group_cpus(group), + &p->cpus_allowed)) continue; - local_group = cpu_isset(this_cpu, group->cpumask); + local_group = cpumask_test_cpu(this_cpu, + sched_group_cpus(group)); /* Tally up the load of all CPUs in the group */ avg_load = 0; - for_each_cpu_mask_nr(i, group->cpumask) { + for_each_cpu(i, sched_group_cpus(group)) { /* Bias balancing toward cpus of our domain */ if (local_group) load = source_load(i, load_idx); @@ -2119,17 +2096,14 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu) * find_idlest_cpu - find the idlest cpu among the cpus in group. */ static int -find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu, - cpumask_t *tmp) +find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu) { unsigned long load, min_load = ULONG_MAX; int idlest = -1; int i; /* Traverse only the allowed CPUs */ - cpus_and(*tmp, group->cpumask, p->cpus_allowed); - - for_each_cpu_mask_nr(i, *tmp) { + for_each_cpu_and(i, sched_group_cpus(group), &p->cpus_allowed) { load = weighted_cpuload(i); if (load < min_load || (load == min_load && i == this_cpu)) { @@ -2171,7 +2145,6 @@ static int sched_balance_self(int cpu, int flag) update_shares(sd); while (sd) { - cpumask_t span, tmpmask; struct sched_group *group; int new_cpu, weight; @@ -2180,14 +2153,13 @@ static int sched_balance_self(int cpu, int flag) continue; } - span = sd->span; group = find_idlest_group(sd, t, cpu); if (!group) { sd = sd->child; continue; } - new_cpu = find_idlest_cpu(group, t, cpu, &tmpmask); + new_cpu = find_idlest_cpu(group, t, cpu); if (new_cpu == -1 || new_cpu == cpu) { /* Now try balancing at a lower domain level of cpu */ sd = sd->child; @@ -2196,10 +2168,10 @@ static int sched_balance_self(int cpu, int flag) /* Now try balancing at a lower domain level of new_cpu */ cpu = new_cpu; + weight = cpumask_weight(sched_domain_span(sd)); sd = NULL; - weight = cpus_weight(span); for_each_domain(cpu, tmp) { - if (weight <= cpus_weight(tmp->span)) + if (weight <= cpumask_weight(sched_domain_span(tmp))) break; if (tmp->flags & flag) sd = tmp; @@ -2244,7 +2216,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync) cpu = task_cpu(p); for_each_domain(this_cpu, sd) { - if (cpu_isset(cpu, sd->span)) { + if (cpumask_test_cpu(cpu, sched_domain_span(sd))) { update_shares(sd); break; } @@ -2292,7 +2264,7 @@ static int try_to_wake_up(struct task_struct *p, unsigned int state, int sync) else { struct sched_domain *sd; for_each_domain(this_cpu, sd) { - if (cpu_isset(cpu, sd->span)) { + if (cpumask_test_cpu(cpu, sched_domain_span(sd))) { schedstat_inc(sd, ttwu_wake_remote); break; } @@ -2838,7 +2810,7 @@ static int double_lock_balance(struct rq *this_rq, struct rq *busiest) return ret; } -static void double_unlock_balance(struct rq *this_rq, struct rq *busiest) +static inline void double_unlock_balance(struct rq *this_rq, struct rq *busiest) __releases(busiest->lock) { spin_unlock(&busiest->lock); @@ -2858,7 +2830,7 @@ static void sched_migrate_task(struct task_struct *p, int dest_cpu) struct rq *rq; rq = task_rq_lock(p, &flags); - if (!cpu_isset(dest_cpu, p->cpus_allowed) + if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed) || unlikely(!cpu_active(dest_cpu))) goto out; @@ -2924,7 +2896,7 @@ int can_migrate_task(struct task_struct *p, struct rq *rq, int this_cpu, * 2) cannot be migrated to this CPU due to cpus_allowed, or * 3) are cache-hot on their current CPU. */ - if (!cpu_isset(this_cpu, p->cpus_allowed)) { + if (!cpumask_test_cpu(this_cpu, &p->cpus_allowed)) { schedstat_inc(p, se.nr_failed_migrations_affine); return 0; } @@ -3099,7 +3071,7 @@ static int move_one_task(struct rq *this_rq, int this_cpu, struct rq *busiest, static struct sched_group * find_busiest_group(struct sched_domain *sd, int this_cpu, unsigned long *imbalance, enum cpu_idle_type idle, - int *sd_idle, const cpumask_t *cpus, int *balance) + int *sd_idle, const struct cpumask *cpus, int *balance) { struct sched_group *busiest = NULL, *this = NULL, *group = sd->groups; unsigned long max_load, avg_load, total_load, this_load, total_pwr; @@ -3135,10 +3107,11 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, unsigned long sum_avg_load_per_task; unsigned long avg_load_per_task; - local_group = cpu_isset(this_cpu, group->cpumask); + local_group = cpumask_test_cpu(this_cpu, + sched_group_cpus(group)); if (local_group) - balance_cpu = first_cpu(group->cpumask); + balance_cpu = cpumask_first(sched_group_cpus(group)); /* Tally up the load of all CPUs in the group */ sum_weighted_load = sum_nr_running = avg_load = 0; @@ -3147,13 +3120,8 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, max_cpu_load = 0; min_cpu_load = ~0UL; - for_each_cpu_mask_nr(i, group->cpumask) { - struct rq *rq; - - if (!cpu_isset(i, *cpus)) - continue; - - rq = cpu_rq(i); + for_each_cpu_and(i, sched_group_cpus(group), cpus) { + struct rq *rq = cpu_rq(i); if (*sd_idle && rq->nr_running) *sd_idle = 0; @@ -3264,8 +3232,8 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, */ if ((sum_nr_running < min_nr_running) || (sum_nr_running == min_nr_running && - first_cpu(group->cpumask) < - first_cpu(group_min->cpumask))) { + cpumask_first(sched_group_cpus(group)) < + cpumask_first(sched_group_cpus(group_min)))) { group_min = group; min_nr_running = sum_nr_running; min_load_per_task = sum_weighted_load / @@ -3280,8 +3248,8 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, if (sum_nr_running <= group_capacity - 1) { if (sum_nr_running > leader_nr_running || (sum_nr_running == leader_nr_running && - first_cpu(group->cpumask) > - first_cpu(group_leader->cpumask))) { + cpumask_first(sched_group_cpus(group)) > + cpumask_first(sched_group_cpus(group_leader)))) { group_leader = group; leader_nr_running = sum_nr_running; } @@ -3420,16 +3388,16 @@ ret: */ static struct rq * find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle, - unsigned long imbalance, const cpumask_t *cpus) + unsigned long imbalance, const struct cpumask *cpus) { struct rq *busiest = NULL, *rq; unsigned long max_load = 0; int i; - for_each_cpu_mask_nr(i, group->cpumask) { + for_each_cpu(i, sched_group_cpus(group)) { unsigned long wl; - if (!cpu_isset(i, *cpus)) + if (!cpumask_test_cpu(i, cpus)) continue; rq = cpu_rq(i); @@ -3459,7 +3427,7 @@ find_busiest_queue(struct sched_group *group, enum cpu_idle_type idle, */ static int load_balance(int this_cpu, struct rq *this_rq, struct sched_domain *sd, enum cpu_idle_type idle, - int *balance, cpumask_t *cpus) + int *balance, struct cpumask *cpus) { int ld_moved, all_pinned = 0, active_balance = 0, sd_idle = 0; struct sched_group *group; @@ -3467,7 +3435,7 @@ static int load_balance(int this_cpu, struct rq *this_rq, struct rq *busiest; unsigned long flags; - cpus_setall(*cpus); + cpumask_setall(cpus); /* * When power savings policy is enabled for the parent domain, idle @@ -3527,8 +3495,8 @@ redo: /* All tasks on this runqueue were pinned by CPU affinity */ if (unlikely(all_pinned)) { - cpu_clear(cpu_of(busiest), *cpus); - if (!cpus_empty(*cpus)) + cpumask_clear_cpu(cpu_of(busiest), cpus); + if (!cpumask_empty(cpus)) goto redo; goto out_balanced; } @@ -3545,7 +3513,8 @@ redo: /* don't kick the migration_thread, if the curr * task on busiest cpu can't be moved to this_cpu */ - if (!cpu_isset(this_cpu, busiest->curr->cpus_allowed)) { + if (!cpumask_test_cpu(this_cpu, + &busiest->curr->cpus_allowed)) { spin_unlock_irqrestore(&busiest->lock, flags); all_pinned = 1; goto out_one_pinned; @@ -3620,7 +3589,7 @@ out: */ static int load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd, - cpumask_t *cpus) + struct cpumask *cpus) { struct sched_group *group; struct rq *busiest = NULL; @@ -3629,7 +3598,7 @@ load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd, int sd_idle = 0; int all_pinned = 0; - cpus_setall(*cpus); + cpumask_setall(cpus); /* * When power savings policy is enabled for the parent domain, idle @@ -3673,8 +3642,8 @@ redo: double_unlock_balance(this_rq, busiest); if (unlikely(all_pinned)) { - cpu_clear(cpu_of(busiest), *cpus); - if (!cpus_empty(*cpus)) + cpumask_clear_cpu(cpu_of(busiest), cpus); + if (!cpumask_empty(cpus)) goto redo; } } @@ -3709,7 +3678,10 @@ static void idle_balance(int this_cpu, struct rq *this_rq) struct sched_domain *sd; int pulled_task = -1; unsigned long next_balance = jiffies + HZ; - cpumask_t tmpmask; + cpumask_var_t tmpmask; + + if (!alloc_cpumask_var(&tmpmask, GFP_ATOMIC)) + return; for_each_domain(this_cpu, sd) { unsigned long interval; @@ -3720,7 +3692,7 @@ static void idle_balance(int this_cpu, struct rq *this_rq) if (sd->flags & SD_BALANCE_NEWIDLE) /* If we've pulled tasks over stop searching: */ pulled_task = load_balance_newidle(this_cpu, this_rq, - sd, &tmpmask); + sd, tmpmask); interval = msecs_to_jiffies(sd->balance_interval); if (time_after(next_balance, sd->last_balance + interval)) @@ -3735,6 +3707,7 @@ static void idle_balance(int this_cpu, struct rq *this_rq) */ this_rq->next_balance = next_balance; } + free_cpumask_var(tmpmask); } /* @@ -3772,7 +3745,7 @@ static void active_load_balance(struct rq *busiest_rq, int busiest_cpu) /* Search for an sd spanning us and the target CPU. */ for_each_domain(target_cpu, sd) { if ((sd->flags & SD_LOAD_BALANCE) && - cpu_isset(busiest_cpu, sd->span)) + cpumask_test_cpu(busiest_cpu, sched_domain_span(sd))) break; } @@ -3791,10 +3764,9 @@ static void active_load_balance(struct rq *busiest_rq, int busiest_cpu) #ifdef CONFIG_NO_HZ static struct { atomic_t load_balancer; - cpumask_t cpu_mask; + cpumask_var_t cpu_mask; } nohz ____cacheline_aligned = { .load_balancer = ATOMIC_INIT(-1), - .cpu_mask = CPU_MASK_NONE, }; /* @@ -3822,7 +3794,7 @@ int select_nohz_load_balancer(int stop_tick) int cpu = smp_processor_id(); if (stop_tick) { - cpu_set(cpu, nohz.cpu_mask); + cpumask_set_cpu(cpu, nohz.cpu_mask); cpu_rq(cpu)->in_nohz_recently = 1; /* @@ -3836,7 +3808,7 @@ int select_nohz_load_balancer(int stop_tick) } /* time for ilb owner also to sleep */ - if (cpus_weight(nohz.cpu_mask) == num_online_cpus()) { + if (cpumask_weight(nohz.cpu_mask) == num_online_cpus()) { if (atomic_read(&nohz.load_balancer) == cpu) atomic_set(&nohz.load_balancer, -1); return 0; @@ -3849,10 +3821,10 @@ int select_nohz_load_balancer(int stop_tick) } else if (atomic_read(&nohz.load_balancer) == cpu) return 1; } else { - if (!cpu_isset(cpu, nohz.cpu_mask)) + if (!cpumask_test_cpu(cpu, nohz.cpu_mask)) return 0; - cpu_clear(cpu, nohz.cpu_mask); + cpumask_clear_cpu(cpu, nohz.cpu_mask); if (atomic_read(&nohz.load_balancer) == cpu) if (atomic_cmpxchg(&nohz.load_balancer, cpu, -1) != cpu) @@ -3880,7 +3852,11 @@ static void rebalance_domains(int cpu, enum cpu_idle_type idle) unsigned long next_balance = jiffies + 60*HZ; int update_next_balance = 0; int need_serialize; - cpumask_t tmp; + cpumask_var_t tmp; + + /* Fails alloc? Rebalancing probably not a priority right now. */ + if (!alloc_cpumask_var(&tmp, GFP_ATOMIC)) + return; for_each_domain(cpu, sd) { if (!(sd->flags & SD_LOAD_BALANCE)) @@ -3905,7 +3881,7 @@ static void rebalance_domains(int cpu, enum cpu_idle_type idle) } if (time_after_eq(jiffies, sd->last_balance + interval)) { - if (load_balance(cpu, rq, sd, idle, &balance, &tmp)) { + if (load_balance(cpu, rq, sd, idle, &balance, tmp)) { /* * We've pulled tasks over so either we're no * longer idle, or one of our SMT siblings is @@ -3939,6 +3915,8 @@ out: */ if (likely(update_next_balance)) rq->next_balance = next_balance; + + free_cpumask_var(tmp); } /* @@ -3963,12 +3941,13 @@ static void run_rebalance_domains(struct softirq_action *h) */ if (this_rq->idle_at_tick && atomic_read(&nohz.load_balancer) == this_cpu) { - cpumask_t cpus = nohz.cpu_mask; struct rq *rq; int balance_cpu; - cpu_clear(this_cpu, cpus); - for_each_cpu_mask_nr(balance_cpu, cpus) { + for_each_cpu(balance_cpu, nohz.cpu_mask) { + if (balance_cpu == this_cpu) + continue; + /* * If this cpu gets work to do, stop the load balancing * work being done for other cpus. Next load @@ -4006,7 +3985,7 @@ static inline void trigger_load_balance(struct rq *rq, int cpu) rq->in_nohz_recently = 0; if (atomic_read(&nohz.load_balancer) == cpu) { - cpu_clear(cpu, nohz.cpu_mask); + cpumask_clear_cpu(cpu, nohz.cpu_mask); atomic_set(&nohz.load_balancer, -1); } @@ -4019,7 +3998,7 @@ static inline void trigger_load_balance(struct rq *rq, int cpu) * TBD: Traverse the sched domains and nominate * the nearest cpu in the nohz.cpu_mask. */ - int ilb = first_cpu(nohz.cpu_mask); + int ilb = cpumask_first(nohz.cpu_mask); if (ilb < nr_cpu_ids) resched_cpu(ilb); @@ -4031,7 +4010,7 @@ static inline void trigger_load_balance(struct rq *rq, int cpu) * cpus with ticks stopped, is it time for that to stop? */ if (rq->idle_at_tick && atomic_read(&nohz.load_balancer) == cpu && - cpus_weight(nohz.cpu_mask) == num_online_cpus()) { + cpumask_weight(nohz.cpu_mask) == num_online_cpus()) { resched_cpu(cpu); return; } @@ -4041,7 +4020,7 @@ static inline void trigger_load_balance(struct rq *rq, int cpu) * someone else, then no need raise the SCHED_SOFTIRQ */ if (rq->idle_at_tick && atomic_read(&nohz.load_balancer) != cpu && - cpu_isset(cpu, nohz.cpu_mask)) + cpumask_test_cpu(cpu, nohz.cpu_mask)) return; #endif if (time_after_eq(jiffies, rq->next_balance)) @@ -4203,7 +4182,6 @@ void account_steal_time(struct task_struct *p, cputime_t steal) if (p == rq->idle) { p->stime = cputime_add(p->stime, steal); - account_group_system_time(p, steal); if (atomic_read(&rq->nr_iowait) > 0) cpustat->iowait = cputime64_add(cpustat->iowait, tmp); else @@ -4339,7 +4317,7 @@ void __kprobes sub_preempt_count(int val) /* * Underflow? */ - if (DEBUG_LOCKS_WARN_ON(val > preempt_count())) + if (DEBUG_LOCKS_WARN_ON(val > preempt_count() - (!!kernel_locked()))) return; /* * Is the spinlock portion underflowing? @@ -5400,10 +5378,9 @@ out_unlock: return retval; } -long sched_setaffinity(pid_t pid, const cpumask_t *in_mask) +long sched_setaffinity(pid_t pid, const struct cpumask *in_mask) { - cpumask_t cpus_allowed; - cpumask_t new_mask = *in_mask; + cpumask_var_t cpus_allowed, new_mask; struct task_struct *p; int retval; @@ -5425,6 +5402,14 @@ long sched_setaffinity(pid_t pid, const cpumask_t *in_mask) get_task_struct(p); read_unlock(&tasklist_lock); + if (!alloc_cpumask_var(&cpus_allowed, GFP_KERNEL)) { + retval = -ENOMEM; + goto out_put_task; + } + if (!alloc_cpumask_var(&new_mask, GFP_KERNEL)) { + retval = -ENOMEM; + goto out_free_cpus_allowed; + } retval = -EPERM; if ((current->euid != p->euid) && (current->euid != p->uid) && !capable(CAP_SYS_NICE)) @@ -5434,37 +5419,41 @@ long sched_setaffinity(pid_t pid, const cpumask_t *in_mask) if (retval) goto out_unlock; - cpuset_cpus_allowed(p, &cpus_allowed); - cpus_and(new_mask, new_mask, cpus_allowed); + cpuset_cpus_allowed(p, cpus_allowed); + cpumask_and(new_mask, in_mask, cpus_allowed); again: - retval = set_cpus_allowed_ptr(p, &new_mask); + retval = set_cpus_allowed_ptr(p, new_mask); if (!retval) { - cpuset_cpus_allowed(p, &cpus_allowed); - if (!cpus_subset(new_mask, cpus_allowed)) { + cpuset_cpus_allowed(p, cpus_allowed); + if (!cpumask_subset(new_mask, cpus_allowed)) { /* * We must have raced with a concurrent cpuset * update. Just reset the cpus_allowed to the * cpuset's cpus_allowed */ - new_mask = cpus_allowed; + cpumask_copy(new_mask, cpus_allowed); goto again; } } out_unlock: + free_cpumask_var(new_mask); +out_free_cpus_allowed: + free_cpumask_var(cpus_allowed); +out_put_task: put_task_struct(p); put_online_cpus(); return retval; } static int get_user_cpu_mask(unsigned long __user *user_mask_ptr, unsigned len, - cpumask_t *new_mask) + struct cpumask *new_mask) { - if (len < sizeof(cpumask_t)) { - memset(new_mask, 0, sizeof(cpumask_t)); - } else if (len > sizeof(cpumask_t)) { - len = sizeof(cpumask_t); - } + if (len < cpumask_size()) + cpumask_clear(new_mask); + else if (len > cpumask_size()) + len = cpumask_size(); + return copy_from_user(new_mask, user_mask_ptr, len) ? -EFAULT : 0; } @@ -5477,17 +5466,20 @@ static int get_user_cpu_mask(unsigned long __user *user_mask_ptr, unsigned len, asmlinkage long sys_sched_setaffinity(pid_t pid, unsigned int len, unsigned long __user *user_mask_ptr) { - cpumask_t new_mask; + cpumask_var_t new_mask; int retval; - retval = get_user_cpu_mask(user_mask_ptr, len, &new_mask); - if (retval) - return retval; + if (!alloc_cpumask_var(&new_mask, GFP_KERNEL)) + return -ENOMEM; - return sched_setaffinity(pid, &new_mask); + retval = get_user_cpu_mask(user_mask_ptr, len, new_mask); + if (retval == 0) + retval = sched_setaffinity(pid, new_mask); + free_cpumask_var(new_mask); + return retval; } -long sched_getaffinity(pid_t pid, cpumask_t *mask) +long sched_getaffinity(pid_t pid, struct cpumask *mask) { struct task_struct *p; int retval; @@ -5504,7 +5496,7 @@ long sched_getaffinity(pid_t pid, cpumask_t *mask) if (retval) goto out_unlock; - cpus_and(*mask, p->cpus_allowed, cpu_online_map); + cpumask_and(mask, &p->cpus_allowed, cpu_online_mask); out_unlock: read_unlock(&tasklist_lock); @@ -5523,19 +5515,24 @@ asmlinkage long sys_sched_getaffinity(pid_t pid, unsigned int len, unsigned long __user *user_mask_ptr) { int ret; - cpumask_t mask; + cpumask_var_t mask; - if (len < sizeof(cpumask_t)) + if (len < cpumask_size()) return -EINVAL; - ret = sched_getaffinity(pid, &mask); - if (ret < 0) - return ret; + if (!alloc_cpumask_var(&mask, GFP_KERNEL)) + return -ENOMEM; - if (copy_to_user(user_mask_ptr, &mask, sizeof(cpumask_t))) - return -EFAULT; + ret = sched_getaffinity(pid, mask); + if (ret == 0) { + if (copy_to_user(user_mask_ptr, mask, cpumask_size())) + ret = -EFAULT; + else + ret = cpumask_size(); + } + free_cpumask_var(mask); - return sizeof(cpumask_t); + return ret; } /** @@ -5877,7 +5874,7 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) idle->se.exec_start = sched_clock(); idle->prio = idle->normal_prio = MAX_PRIO; - idle->cpus_allowed = cpumask_of_cpu(cpu); + cpumask_copy(&idle->cpus_allowed, cpumask_of(cpu)); __set_task_cpu(idle, cpu); rq->curr = rq->idle = idle; @@ -5896,6 +5893,7 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) * The idle tasks have their own, simple scheduling class: */ idle->sched_class = &idle_sched_class; + ftrace_retfunc_init_task(idle); } /* @@ -5903,9 +5901,9 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) * indicates which cpus entered this state. This is used * in the rcu update to wait only for active cpus. For system * which do not switch off the HZ timer nohz_cpu_mask should - * always be CPU_MASK_NONE. + * always be CPU_BITS_NONE. */ -cpumask_t nohz_cpu_mask = CPU_MASK_NONE; +cpumask_var_t nohz_cpu_mask; /* * Increase the granularity value when there are more CPUs, @@ -5960,7 +5958,7 @@ static inline void sched_init_granularity(void) * task must not exit() & deallocate itself prematurely. The * call is not atomic; no spinlocks may be held. */ -int set_cpus_allowed_ptr(struct task_struct *p, const cpumask_t *new_mask) +int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask) { struct migration_req req; unsigned long flags; @@ -5968,13 +5966,13 @@ int set_cpus_allowed_ptr(struct task_struct *p, const cpumask_t *new_mask) int ret = 0; rq = task_rq_lock(p, &flags); - if (!cpus_intersects(*new_mask, cpu_online_map)) { + if (!cpumask_intersects(new_mask, cpu_online_mask)) { ret = -EINVAL; goto out; } if (unlikely((p->flags & PF_THREAD_BOUND) && p != current && - !cpus_equal(p->cpus_allowed, *new_mask))) { + !cpumask_equal(&p->cpus_allowed, new_mask))) { ret = -EINVAL; goto out; } @@ -5982,15 +5980,15 @@ int set_cpus_allowed_ptr(struct task_struct *p, const cpumask_t *new_mask) if (p->sched_class->set_cpus_allowed) p->sched_class->set_cpus_allowed(p, new_mask); else { - p->cpus_allowed = *new_mask; - p->rt.nr_cpus_allowed = cpus_weight(*new_mask); + cpumask_copy(&p->cpus_allowed, new_mask); + p->rt.nr_cpus_allowed = cpumask_weight(new_mask); } /* Can the task run on the task's current CPU? If so, we're done */ - if (cpu_isset(task_cpu(p), *new_mask)) + if (cpumask_test_cpu(task_cpu(p), new_mask)) goto out; - if (migrate_task(p, any_online_cpu(*new_mask), &req)) { + if (migrate_task(p, cpumask_any_and(cpu_online_mask, new_mask), &req)) { /* Need help from migration thread: drop lock and wait. */ task_rq_unlock(rq, &flags); wake_up_process(rq->migration_thread); @@ -6032,7 +6030,7 @@ static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu) if (task_cpu(p) != src_cpu) goto done; /* Affinity changed (again). */ - if (!cpu_isset(dest_cpu, p->cpus_allowed)) + if (!cpumask_test_cpu(dest_cpu, &p->cpus_allowed)) goto fail; on_rq = p->se.on_rq; @@ -6126,54 +6124,46 @@ static int __migrate_task_irq(struct task_struct *p, int src_cpu, int dest_cpu) /* * Figure out where task on dead CPU should go, use force if necessary. - * NOTE: interrupts should be disabled by the caller */ static void move_task_off_dead_cpu(int dead_cpu, struct task_struct *p) { - unsigned long flags; - cpumask_t mask; - struct rq *rq; int dest_cpu; + /* FIXME: Use cpumask_of_node here. */ + cpumask_t _nodemask = node_to_cpumask(cpu_to_node(dead_cpu)); + const struct cpumask *nodemask = &_nodemask; + +again: + /* Look for allowed, online CPU in same node. */ + for_each_cpu_and(dest_cpu, nodemask, cpu_online_mask) + if (cpumask_test_cpu(dest_cpu, &p->cpus_allowed)) + goto move; + + /* Any allowed, online CPU? */ + dest_cpu = cpumask_any_and(&p->cpus_allowed, cpu_online_mask); + if (dest_cpu < nr_cpu_ids) + goto move; + + /* No more Mr. Nice Guy. */ + if (dest_cpu >= nr_cpu_ids) { + cpuset_cpus_allowed_locked(p, &p->cpus_allowed); + dest_cpu = cpumask_any_and(cpu_online_mask, &p->cpus_allowed); - do { - /* On same node? */ - mask = node_to_cpumask(cpu_to_node(dead_cpu)); - cpus_and(mask, mask, p->cpus_allowed); - dest_cpu = any_online_cpu(mask); - - /* On any allowed CPU? */ - if (dest_cpu >= nr_cpu_ids) - dest_cpu = any_online_cpu(p->cpus_allowed); - - /* No more Mr. Nice Guy. */ - if (dest_cpu >= nr_cpu_ids) { - cpumask_t cpus_allowed; - - cpuset_cpus_allowed_locked(p, &cpus_allowed); - /* - * Try to stay on the same cpuset, where the - * current cpuset may be a subset of all cpus. - * The cpuset_cpus_allowed_locked() variant of - * cpuset_cpus_allowed() will not block. It must be - * called within calls to cpuset_lock/cpuset_unlock. - */ - rq = task_rq_lock(p, &flags); - p->cpus_allowed = cpus_allowed; - dest_cpu = any_online_cpu(p->cpus_allowed); - task_rq_unlock(rq, &flags); - - /* - * Don't tell them about moving exiting tasks or - * kernel threads (both mm NULL), since they never - * leave kernel. - */ - if (p->mm && printk_ratelimit()) { - printk(KERN_INFO "process %d (%s) no " - "longer affine to cpu%d\n", - task_pid_nr(p), p->comm, dead_cpu); - } + /* + * Don't tell them about moving exiting tasks or + * kernel threads (both mm NULL), since they never + * leave kernel. + */ + if (p->mm && printk_ratelimit()) { + printk(KERN_INFO "process %d (%s) no " + "longer affine to cpu%d\n", + task_pid_nr(p), p->comm, dead_cpu); } - } while (!__migrate_task_irq(p, dead_cpu, dest_cpu)); + } + +move: + /* It can have affinity changed while we were choosing. */ + if (unlikely(!__migrate_task_irq(p, dead_cpu, dest_cpu))) + goto again; } /* @@ -6185,7 +6175,7 @@ static void move_task_off_dead_cpu(int dead_cpu, struct task_struct *p) */ static void migrate_nr_uninterruptible(struct rq *rq_src) { - struct rq *rq_dest = cpu_rq(any_online_cpu(*CPU_MASK_ALL_PTR)); + struct rq *rq_dest = cpu_rq(cpumask_any(cpu_online_mask)); unsigned long flags; local_irq_save(flags); @@ -6475,7 +6465,7 @@ static void set_rq_online(struct rq *rq) if (!rq->online) { const struct sched_class *class; - cpu_set(rq->cpu, rq->rd->online); + cpumask_set_cpu(rq->cpu, rq->rd->online); rq->online = 1; for_each_class(class) { @@ -6495,7 +6485,7 @@ static void set_rq_offline(struct rq *rq) class->rq_offline(rq); } - cpu_clear(rq->cpu, rq->rd->online); + cpumask_clear_cpu(rq->cpu, rq->rd->online); rq->online = 0; } } @@ -6536,7 +6526,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) rq = cpu_rq(cpu); spin_lock_irqsave(&rq->lock, flags); if (rq->rd) { - BUG_ON(!cpu_isset(cpu, rq->rd->span)); + BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); set_rq_online(rq); } @@ -6550,7 +6540,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) break; /* Unbind it from offline cpu so it can run. Fall thru. */ kthread_bind(cpu_rq(cpu)->migration_thread, - any_online_cpu(cpu_online_map)); + cpumask_any(cpu_online_mask)); kthread_stop(cpu_rq(cpu)->migration_thread); cpu_rq(cpu)->migration_thread = NULL; break; @@ -6598,7 +6588,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) rq = cpu_rq(cpu); spin_lock_irqsave(&rq->lock, flags); if (rq->rd) { - BUG_ON(!cpu_isset(cpu, rq->rd->span)); + BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); set_rq_offline(rq); } spin_unlock_irqrestore(&rq->lock, flags); @@ -6636,36 +6626,14 @@ early_initcall(migration_init); #ifdef CONFIG_SCHED_DEBUG -static inline const char *sd_level_to_string(enum sched_domain_level lvl) -{ - switch (lvl) { - case SD_LV_NONE: - return "NONE"; - case SD_LV_SIBLING: - return "SIBLING"; - case SD_LV_MC: - return "MC"; - case SD_LV_CPU: - return "CPU"; - case SD_LV_NODE: - return "NODE"; - case SD_LV_ALLNODES: - return "ALLNODES"; - case SD_LV_MAX: - return "MAX"; - - } - return "MAX"; -} - static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, - cpumask_t *groupmask) + struct cpumask *groupmask) { struct sched_group *group = sd->groups; char str[256]; - cpulist_scnprintf(str, sizeof(str), sd->span); - cpus_clear(*groupmask); + cpulist_scnprintf(str, sizeof(str), *sched_domain_span(sd)); + cpumask_clear(groupmask); printk(KERN_DEBUG "%*s domain %d: ", level, "", level); @@ -6677,14 +6645,13 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, return -1; } - printk(KERN_CONT "span %s level %s\n", - str, sd_level_to_string(sd->level)); + printk(KERN_CONT "span %s level %s\n", str, sd->name); - if (!cpu_isset(cpu, sd->span)) { + if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) { printk(KERN_ERR "ERROR: domain->span does not contain " "CPU%d\n", cpu); } - if (!cpu_isset(cpu, group->cpumask)) { + if (!cpumask_test_cpu(cpu, sched_group_cpus(group))) { printk(KERN_ERR "ERROR: domain->groups does not contain" " CPU%d\n", cpu); } @@ -6704,31 +6671,32 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, break; } - if (!cpus_weight(group->cpumask)) { + if (!cpumask_weight(sched_group_cpus(group))) { printk(KERN_CONT "\n"); printk(KERN_ERR "ERROR: empty group\n"); break; } - if (cpus_intersects(*groupmask, group->cpumask)) { + if (cpumask_intersects(groupmask, sched_group_cpus(group))) { printk(KERN_CONT "\n"); printk(KERN_ERR "ERROR: repeated CPUs\n"); break; } - cpus_or(*groupmask, *groupmask, group->cpumask); + cpumask_or(groupmask, groupmask, sched_group_cpus(group)); - cpulist_scnprintf(str, sizeof(str), group->cpumask); + cpulist_scnprintf(str, sizeof(str), *sched_group_cpus(group)); printk(KERN_CONT " %s", str); group = group->next; } while (group != sd->groups); printk(KERN_CONT "\n"); - if (!cpus_equal(sd->span, *groupmask)) + if (!cpumask_equal(sched_domain_span(sd), groupmask)) printk(KERN_ERR "ERROR: groups don't span domain->span\n"); - if (sd->parent && !cpus_subset(*groupmask, sd->parent->span)) + if (sd->parent && + !cpumask_subset(groupmask, sched_domain_span(sd->parent))) printk(KERN_ERR "ERROR: parent span is not a superset " "of domain->span\n"); return 0; @@ -6736,7 +6704,7 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, static void sched_domain_debug(struct sched_domain *sd, int cpu) { - cpumask_t *groupmask; + cpumask_var_t groupmask; int level = 0; if (!sd) { @@ -6746,8 +6714,7 @@ static void sched_domain_debug(struct sched_domain *sd, int cpu) printk(KERN_DEBUG "CPU%d attaching sched-domain:\n", cpu); - groupmask = kmalloc(sizeof(cpumask_t), GFP_KERNEL); - if (!groupmask) { + if (!alloc_cpumask_var(&groupmask, GFP_KERNEL)) { printk(KERN_DEBUG "Cannot load-balance (out of memory)\n"); return; } @@ -6760,7 +6727,7 @@ static void sched_domain_debug(struct sched_domain *sd, int cpu) if (!sd) break; } - kfree(groupmask); + free_cpumask_var(groupmask); } #else /* !CONFIG_SCHED_DEBUG */ # define sched_domain_debug(sd, cpu) do { } while (0) @@ -6768,7 +6735,7 @@ static void sched_domain_debug(struct sched_domain *sd, int cpu) static int sd_degenerate(struct sched_domain *sd) { - if (cpus_weight(sd->span) == 1) + if (cpumask_weight(sched_domain_span(sd)) == 1) return 1; /* Following flags need at least 2 groups */ @@ -6799,7 +6766,7 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent) if (sd_degenerate(parent)) return 1; - if (!cpus_equal(sd->span, parent->span)) + if (!cpumask_equal(sched_domain_span(sd), sched_domain_span(parent))) return 0; /* Does parent contain flags not in child? */ @@ -6821,6 +6788,16 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent) return 1; } +static void free_rootdomain(struct root_domain *rd) +{ + cpupri_cleanup(&rd->cpupri); + + free_cpumask_var(rd->rto_mask); + free_cpumask_var(rd->online); + free_cpumask_var(rd->span); + kfree(rd); +} + static void rq_attach_root(struct rq *rq, struct root_domain *rd) { unsigned long flags; @@ -6830,38 +6807,63 @@ static void rq_attach_root(struct rq *rq, struct root_domain *rd) if (rq->rd) { struct root_domain *old_rd = rq->rd; - if (cpu_isset(rq->cpu, old_rd->online)) + if (cpumask_test_cpu(rq->cpu, old_rd->online)) set_rq_offline(rq); - cpu_clear(rq->cpu, old_rd->span); + cpumask_clear_cpu(rq->cpu, old_rd->span); if (atomic_dec_and_test(&old_rd->refcount)) - kfree(old_rd); + free_rootdomain(old_rd); } atomic_inc(&rd->refcount); rq->rd = rd; - cpu_set(rq->cpu, rd->span); - if (cpu_isset(rq->cpu, cpu_online_map)) + cpumask_set_cpu(rq->cpu, rd->span); + if (cpumask_test_cpu(rq->cpu, cpu_online_mask)) set_rq_online(rq); spin_unlock_irqrestore(&rq->lock, flags); } -static void init_rootdomain(struct root_domain *rd) +static int init_rootdomain(struct root_domain *rd, bool bootmem) { memset(rd, 0, sizeof(*rd)); - cpus_clear(rd->span); - cpus_clear(rd->online); + if (bootmem) { + alloc_bootmem_cpumask_var(&def_root_domain.span); + alloc_bootmem_cpumask_var(&def_root_domain.online); + alloc_bootmem_cpumask_var(&def_root_domain.rto_mask); + cpupri_init(&rd->cpupri, true); + return 0; + } + + if (!alloc_cpumask_var(&rd->span, GFP_KERNEL)) + goto free_rd; + if (!alloc_cpumask_var(&rd->online, GFP_KERNEL)) + goto free_span; + if (!alloc_cpumask_var(&rd->rto_mask, GFP_KERNEL)) + goto free_online; + + if (cpupri_init(&rd->cpupri, false) != 0) + goto free_rto_mask; + return 0; - cpupri_init(&rd->cpupri); +free_rto_mask: + free_cpumask_var(rd->rto_mask); +free_online: + free_cpumask_var(rd->online); +free_span: + free_cpumask_var(rd->span); +free_rd: + kfree(rd); + return -ENOMEM; } static void init_defrootdomain(void) { - init_rootdomain(&def_root_domain); + init_rootdomain(&def_root_domain, true); + atomic_set(&def_root_domain.refcount, 1); } @@ -6873,7 +6875,10 @@ static struct root_domain *alloc_rootdomain(void) if (!rd) return NULL; - init_rootdomain(rd); + if (init_rootdomain(rd, false) != 0) { + kfree(rd); + return NULL; + } return rd; } @@ -6915,19 +6920,12 @@ cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu) } /* cpus with isolated domains */ -static cpumask_t cpu_isolated_map = CPU_MASK_NONE; +static cpumask_var_t cpu_isolated_map; /* Setup the mask of cpus configured for isolated domains */ static int __init isolated_cpu_setup(char *str) { - static int __initdata ints[NR_CPUS]; - int i; - - str = get_options(str, ARRAY_SIZE(ints), ints); - cpus_clear(cpu_isolated_map); - for (i = 1; i <= ints[0]; i++) - if (ints[i] < NR_CPUS) - cpu_set(ints[i], cpu_isolated_map); + cpulist_parse(str, *cpu_isolated_map); return 1; } @@ -6936,42 +6934,43 @@ __setup("isolcpus=", isolated_cpu_setup); /* * init_sched_build_groups takes the cpumask we wish to span, and a pointer * to a function which identifies what group(along with sched group) a CPU - * belongs to. The return value of group_fn must be a >= 0 and < NR_CPUS - * (due to the fact that we keep track of groups covered with a cpumask_t). + * belongs to. The return value of group_fn must be a >= 0 and < nr_cpu_ids + * (due to the fact that we keep track of groups covered with a struct cpumask). * * init_sched_build_groups will build a circular linked list of the groups * covered by the given span, and will set each group's ->cpumask correctly, * and ->cpu_power to 0. */ static void -init_sched_build_groups(const cpumask_t *span, const cpumask_t *cpu_map, - int (*group_fn)(int cpu, const cpumask_t *cpu_map, +init_sched_build_groups(const struct cpumask *span, + const struct cpumask *cpu_map, + int (*group_fn)(int cpu, const struct cpumask *cpu_map, struct sched_group **sg, - cpumask_t *tmpmask), - cpumask_t *covered, cpumask_t *tmpmask) + struct cpumask *tmpmask), + struct cpumask *covered, struct cpumask *tmpmask) { struct sched_group *first = NULL, *last = NULL; int i; - cpus_clear(*covered); + cpumask_clear(covered); - for_each_cpu_mask_nr(i, *span) { + for_each_cpu(i, span) { struct sched_group *sg; int group = group_fn(i, cpu_map, &sg, tmpmask); int j; - if (cpu_isset(i, *covered)) + if (cpumask_test_cpu(i, covered)) continue; - cpus_clear(sg->cpumask); + cpumask_clear(sched_group_cpus(sg)); sg->__cpu_power = 0; - for_each_cpu_mask_nr(j, *span) { + for_each_cpu(j, span) { if (group_fn(j, cpu_map, NULL, tmpmask) != group) continue; - cpu_set(j, *covered); - cpu_set(j, sg->cpumask); + cpumask_set_cpu(j, covered); + cpumask_set_cpu(j, sched_group_cpus(sg)); } if (!first) first = sg; @@ -7035,9 +7034,10 @@ static int find_next_best_node(int node, nodemask_t *used_nodes) * should be one that prevents unnecessary balancing, but also spreads tasks * out optimally. */ -static void sched_domain_node_span(int node, cpumask_t *span) +static void sched_domain_node_span(int node, struct cpumask *span) { nodemask_t used_nodes; + /* FIXME: use cpumask_of_node() */ node_to_cpumask_ptr(nodemask, node); int i; @@ -7059,18 +7059,33 @@ static void sched_domain_node_span(int node, cpumask_t *span) int sched_smt_power_savings = 0, sched_mc_power_savings = 0; /* + * The cpus mask in sched_group and sched_domain hangs off the end. + * FIXME: use cpumask_var_t or dynamic percpu alloc to avoid wasting space + * for nr_cpu_ids < CONFIG_NR_CPUS. + */ +struct static_sched_group { + struct sched_group sg; + DECLARE_BITMAP(cpus, CONFIG_NR_CPUS); +}; + +struct static_sched_domain { + struct sched_domain sd; + DECLARE_BITMAP(span, CONFIG_NR_CPUS); +}; + +/* * SMT sched-domains: */ #ifdef CONFIG_SCHED_SMT -static DEFINE_PER_CPU(struct sched_domain, cpu_domains); -static DEFINE_PER_CPU(struct sched_group, sched_group_cpus); +static DEFINE_PER_CPU(struct static_sched_domain, cpu_domains); +static DEFINE_PER_CPU(struct static_sched_group, sched_group_cpus); static int -cpu_to_cpu_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg, - cpumask_t *unused) +cpu_to_cpu_group(int cpu, const struct cpumask *cpu_map, + struct sched_group **sg, struct cpumask *unused) { if (sg) - *sg = &per_cpu(sched_group_cpus, cpu); + *sg = &per_cpu(sched_group_cpus, cpu).sg; return cpu; } #endif /* CONFIG_SCHED_SMT */ @@ -7079,56 +7094,55 @@ cpu_to_cpu_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg, * multi-core sched-domains: */ #ifdef CONFIG_SCHED_MC -static DEFINE_PER_CPU(struct sched_domain, core_domains); -static DEFINE_PER_CPU(struct sched_group, sched_group_core); +static DEFINE_PER_CPU(struct static_sched_domain, core_domains); +static DEFINE_PER_CPU(struct static_sched_group, sched_group_core); #endif /* CONFIG_SCHED_MC */ #if defined(CONFIG_SCHED_MC) && defined(CONFIG_SCHED_SMT) static int -cpu_to_core_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg, - cpumask_t *mask) +cpu_to_core_group(int cpu, const struct cpumask *cpu_map, + struct sched_group **sg, struct cpumask *mask) { int group; - *mask = per_cpu(cpu_sibling_map, cpu); - cpus_and(*mask, *mask, *cpu_map); - group = first_cpu(*mask); + cpumask_and(mask, &per_cpu(cpu_sibling_map, cpu), cpu_map); + group = cpumask_first(mask); if (sg) - *sg = &per_cpu(sched_group_core, group); + *sg = &per_cpu(sched_group_core, group).sg; return group; } #elif defined(CONFIG_SCHED_MC) static int -cpu_to_core_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg, - cpumask_t *unused) +cpu_to_core_group(int cpu, const struct cpumask *cpu_map, + struct sched_group **sg, struct cpumask *unused) { if (sg) - *sg = &per_cpu(sched_group_core, cpu); + *sg = &per_cpu(sched_group_core, cpu).sg; return cpu; } #endif -static DEFINE_PER_CPU(struct sched_domain, phys_domains); -static DEFINE_PER_CPU(struct sched_group, sched_group_phys); +static DEFINE_PER_CPU(struct static_sched_domain, phys_domains); +static DEFINE_PER_CPU(struct static_sched_group, sched_group_phys); static int -cpu_to_phys_group(int cpu, const cpumask_t *cpu_map, struct sched_group **sg, - cpumask_t *mask) +cpu_to_phys_group(int cpu, const struct cpumask *cpu_map, + struct sched_group **sg, struct cpumask *mask) { int group; #ifdef CONFIG_SCHED_MC + /* FIXME: Use cpu_coregroup_mask. */ *mask = cpu_coregroup_map(cpu); cpus_and(*mask, *mask, *cpu_map); - group = first_cpu(*mask); + group = cpumask_first(mask); #elif defined(CONFIG_SCHED_SMT) - *mask = per_cpu(cpu_sibling_map, cpu); - cpus_and(*mask, *mask, *cpu_map); - group = first_cpu(*mask); + cpumask_and(mask, &per_cpu(cpu_sibling_map, cpu), cpu_map); + group = cpumask_first(mask); #else group = cpu; #endif if (sg) - *sg = &per_cpu(sched_group_phys, group); + *sg = &per_cpu(sched_group_phys, group).sg; return group; } @@ -7142,19 +7156,21 @@ static DEFINE_PER_CPU(struct sched_domain, node_domains); static struct sched_group ***sched_group_nodes_bycpu; static DEFINE_PER_CPU(struct sched_domain, allnodes_domains); -static DEFINE_PER_CPU(struct sched_group, sched_group_allnodes); +static DEFINE_PER_CPU(struct static_sched_group, sched_group_allnodes); -static int cpu_to_allnodes_group(int cpu, const cpumask_t *cpu_map, - struct sched_group **sg, cpumask_t *nodemask) +static int cpu_to_allnodes_group(int cpu, const struct cpumask *cpu_map, + struct sched_group **sg, + struct cpumask *nodemask) { int group; + /* FIXME: use cpumask_of_node */ + node_to_cpumask_ptr(pnodemask, cpu_to_node(cpu)); - *nodemask = node_to_cpumask(cpu_to_node(cpu)); - cpus_and(*nodemask, *nodemask, *cpu_map); - group = first_cpu(*nodemask); + cpumask_and(nodemask, pnodemask, cpu_map); + group = cpumask_first(nodemask); if (sg) - *sg = &per_cpu(sched_group_allnodes, group); + *sg = &per_cpu(sched_group_allnodes, group).sg; return group; } @@ -7166,11 +7182,11 @@ static void init_numa_sched_groups_power(struct sched_group *group_head) if (!sg) return; do { - for_each_cpu_mask_nr(j, sg->cpumask) { + for_each_cpu(j, sched_group_cpus(sg)) { struct sched_domain *sd; - sd = &per_cpu(phys_domains, j); - if (j != first_cpu(sd->groups->cpumask)) { + sd = &per_cpu(phys_domains, j).sd; + if (j != cpumask_first(sched_group_cpus(sd->groups))) { /* * Only add "power" once for each * physical package. @@ -7187,11 +7203,12 @@ static void init_numa_sched_groups_power(struct sched_group *group_head) #ifdef CONFIG_NUMA /* Free memory allocated for various sched_group structures */ -static void free_sched_groups(const cpumask_t *cpu_map, cpumask_t *nodemask) +static void free_sched_groups(const struct cpumask *cpu_map, + struct cpumask *nodemask) { int cpu, i; - for_each_cpu_mask_nr(cpu, *cpu_map) { + for_each_cpu(cpu, cpu_map) { struct sched_group **sched_group_nodes = sched_group_nodes_bycpu[cpu]; @@ -7200,10 +7217,11 @@ static void free_sched_groups(const cpumask_t *cpu_map, cpumask_t *nodemask) for (i = 0; i < nr_node_ids; i++) { struct sched_group *oldsg, *sg = sched_group_nodes[i]; + /* FIXME: Use cpumask_of_node */ + node_to_cpumask_ptr(pnodemask, i); - *nodemask = node_to_cpumask(i); - cpus_and(*nodemask, *nodemask, *cpu_map); - if (cpus_empty(*nodemask)) + cpus_and(*nodemask, *pnodemask, *cpu_map); + if (cpumask_empty(nodemask)) continue; if (sg == NULL) @@ -7221,7 +7239,8 @@ next_sg: } } #else /* !CONFIG_NUMA */ -static void free_sched_groups(const cpumask_t *cpu_map, cpumask_t *nodemask) +static void free_sched_groups(const struct cpumask *cpu_map, + struct cpumask *nodemask) { } #endif /* CONFIG_NUMA */ @@ -7247,7 +7266,7 @@ static void init_sched_groups_power(int cpu, struct sched_domain *sd) WARN_ON(!sd || !sd->groups); - if (cpu != first_cpu(sd->groups->cpumask)) + if (cpu != cpumask_first(sched_group_cpus(sd->groups))) return; child = sd->child; @@ -7312,40 +7331,6 @@ SD_INIT_FUNC(CPU) SD_INIT_FUNC(MC) #endif -/* - * To minimize stack usage kmalloc room for cpumasks and share the - * space as the usage in build_sched_domains() dictates. Used only - * if the amount of space is significant. - */ -struct allmasks { - cpumask_t tmpmask; /* make this one first */ - union { - cpumask_t nodemask; - cpumask_t this_sibling_map; - cpumask_t this_core_map; - }; - cpumask_t send_covered; - -#ifdef CONFIG_NUMA - cpumask_t domainspan; - cpumask_t covered; - cpumask_t notcovered; -#endif -}; - -#if NR_CPUS > 128 -#define SCHED_CPUMASK_ALLOC 1 -#define SCHED_CPUMASK_FREE(v) kfree(v) -#define SCHED_CPUMASK_DECLARE(v) struct allmasks *v -#else -#define SCHED_CPUMASK_ALLOC 0 -#define SCHED_CPUMASK_FREE(v) -#define SCHED_CPUMASK_DECLARE(v) struct allmasks _v, *v = &_v -#endif - -#define SCHED_CPUMASK_VAR(v, a) cpumask_t *v = (cpumask_t *) \ - ((unsigned long)(a) + offsetof(struct allmasks, v)) - static int default_relax_domain_level = -1; static int __init setup_relax_domain_level(char *str) @@ -7385,17 +7370,38 @@ static void set_domain_attribute(struct sched_domain *sd, * Build sched domains for a given set of cpus and attach the sched domains * to the individual cpus */ -static int __build_sched_domains(const cpumask_t *cpu_map, +static int __build_sched_domains(const struct cpumask *cpu_map, struct sched_domain_attr *attr) { - int i; + int i, err = -ENOMEM; struct root_domain *rd; - SCHED_CPUMASK_DECLARE(allmasks); - cpumask_t *tmpmask; + cpumask_var_t nodemask, this_sibling_map, this_core_map, send_covered, + tmpmask; #ifdef CONFIG_NUMA + cpumask_var_t domainspan, covered, notcovered; struct sched_group **sched_group_nodes = NULL; int sd_allnodes = 0; + if (!alloc_cpumask_var(&domainspan, GFP_KERNEL)) + goto out; + if (!alloc_cpumask_var(&covered, GFP_KERNEL)) + goto free_domainspan; + if (!alloc_cpumask_var(¬covered, GFP_KERNEL)) + goto free_covered; +#endif + + if (!alloc_cpumask_var(&nodemask, GFP_KERNEL)) + goto free_notcovered; + if (!alloc_cpumask_var(&this_sibling_map, GFP_KERNEL)) + goto free_nodemask; + if (!alloc_cpumask_var(&this_core_map, GFP_KERNEL)) + goto free_this_sibling_map; + if (!alloc_cpumask_var(&send_covered, GFP_KERNEL)) + goto free_this_core_map; + if (!alloc_cpumask_var(&tmpmask, GFP_KERNEL)) + goto free_send_covered; + +#ifdef CONFIG_NUMA /* * Allocate the per-node list of sched groups */ @@ -7403,55 +7409,37 @@ static int __build_sched_domains(const cpumask_t *cpu_map, GFP_KERNEL); if (!sched_group_nodes) { printk(KERN_WARNING "Can not alloc sched group node list\n"); - return -ENOMEM; + goto free_tmpmask; } #endif rd = alloc_rootdomain(); if (!rd) { printk(KERN_WARNING "Cannot alloc root domain\n"); -#ifdef CONFIG_NUMA - kfree(sched_group_nodes); -#endif - return -ENOMEM; - } - -#if SCHED_CPUMASK_ALLOC - /* get space for all scratch cpumask variables */ - allmasks = kmalloc(sizeof(*allmasks), GFP_KERNEL); - if (!allmasks) { - printk(KERN_WARNING "Cannot alloc cpumask array\n"); - kfree(rd); -#ifdef CONFIG_NUMA - kfree(sched_group_nodes); -#endif - return -ENOMEM; + goto free_sched_groups; } -#endif - tmpmask = (cpumask_t *)allmasks; - #ifdef CONFIG_NUMA - sched_group_nodes_bycpu[first_cpu(*cpu_map)] = sched_group_nodes; + sched_group_nodes_bycpu[cpumask_first(cpu_map)] = sched_group_nodes; #endif /* * Set up domains for cpus specified by the cpu_map. */ - for_each_cpu_mask_nr(i, *cpu_map) { + for_each_cpu(i, cpu_map) { struct sched_domain *sd = NULL, *p; - SCHED_CPUMASK_VAR(nodemask, allmasks); + /* FIXME: use cpumask_of_node */ *nodemask = node_to_cpumask(cpu_to_node(i)); cpus_and(*nodemask, *nodemask, *cpu_map); #ifdef CONFIG_NUMA - if (cpus_weight(*cpu_map) > - SD_NODES_PER_DOMAIN*cpus_weight(*nodemask)) { + if (cpumask_weight(cpu_map) > + SD_NODES_PER_DOMAIN*cpumask_weight(nodemask)) { sd = &per_cpu(allnodes_domains, i); SD_INIT(sd, ALLNODES); set_domain_attribute(sd, attr); - sd->span = *cpu_map; + cpumask_copy(sched_domain_span(sd), cpu_map); cpu_to_allnodes_group(i, cpu_map, &sd->groups, tmpmask); p = sd; sd_allnodes = 1; @@ -7461,18 +7449,19 @@ static int __build_sched_domains(const cpumask_t *cpu_map, sd = &per_cpu(node_domains, i); SD_INIT(sd, NODE); set_domain_attribute(sd, attr); - sched_domain_node_span(cpu_to_node(i), &sd->span); + sched_domain_node_span(cpu_to_node(i), sched_domain_span(sd)); sd->parent = p; if (p) p->child = sd; - cpus_and(sd->span, sd->span, *cpu_map); + cpumask_and(sched_domain_span(sd), + sched_domain_span(sd), cpu_map); #endif p = sd; - sd = &per_cpu(phys_domains, i); + sd = &per_cpu(phys_domains, i).sd; SD_INIT(sd, CPU); set_domain_attribute(sd, attr); - sd->span = *nodemask; + cpumask_copy(sched_domain_span(sd), nodemask); sd->parent = p; if (p) p->child = sd; @@ -7480,11 +7469,12 @@ static int __build_sched_domains(const cpumask_t *cpu_map, #ifdef CONFIG_SCHED_MC p = sd; - sd = &per_cpu(core_domains, i); + sd = &per_cpu(core_domains, i).sd; SD_INIT(sd, MC); set_domain_attribute(sd, attr); - sd->span = cpu_coregroup_map(i); - cpus_and(sd->span, sd->span, *cpu_map); + *sched_domain_span(sd) = cpu_coregroup_map(i); + cpumask_and(sched_domain_span(sd), + sched_domain_span(sd), cpu_map); sd->parent = p; p->child = sd; cpu_to_core_group(i, cpu_map, &sd->groups, tmpmask); @@ -7492,11 +7482,11 @@ static int __build_sched_domains(const cpumask_t *cpu_map, #ifdef CONFIG_SCHED_SMT p = sd; - sd = &per_cpu(cpu_domains, i); + sd = &per_cpu(cpu_domains, i).sd; SD_INIT(sd, SIBLING); set_domain_attribute(sd, attr); - sd->span = per_cpu(cpu_sibling_map, i); - cpus_and(sd->span, sd->span, *cpu_map); + cpumask_and(sched_domain_span(sd), + &per_cpu(cpu_sibling_map, i), cpu_map); sd->parent = p; p->child = sd; cpu_to_cpu_group(i, cpu_map, &sd->groups, tmpmask); @@ -7505,13 +7495,10 @@ static int __build_sched_domains(const cpumask_t *cpu_map, #ifdef CONFIG_SCHED_SMT /* Set up CPU (sibling) groups */ - for_each_cpu_mask_nr(i, *cpu_map) { - SCHED_CPUMASK_VAR(this_sibling_map, allmasks); - SCHED_CPUMASK_VAR(send_covered, allmasks); - - *this_sibling_map = per_cpu(cpu_sibling_map, i); - cpus_and(*this_sibling_map, *this_sibling_map, *cpu_map); - if (i != first_cpu(*this_sibling_map)) + for_each_cpu(i, cpu_map) { + cpumask_and(this_sibling_map, + &per_cpu(cpu_sibling_map, i), cpu_map); + if (i != cpumask_first(this_sibling_map)) continue; init_sched_build_groups(this_sibling_map, cpu_map, @@ -7522,13 +7509,11 @@ static int __build_sched_domains(const cpumask_t *cpu_map, #ifdef CONFIG_SCHED_MC /* Set up multi-core groups */ - for_each_cpu_mask_nr(i, *cpu_map) { - SCHED_CPUMASK_VAR(this_core_map, allmasks); - SCHED_CPUMASK_VAR(send_covered, allmasks); - + for_each_cpu(i, cpu_map) { + /* FIXME: Use cpu_coregroup_mask */ *this_core_map = cpu_coregroup_map(i); cpus_and(*this_core_map, *this_core_map, *cpu_map); - if (i != first_cpu(*this_core_map)) + if (i != cpumask_first(this_core_map)) continue; init_sched_build_groups(this_core_map, cpu_map, @@ -7539,12 +7524,10 @@ static int __build_sched_domains(const cpumask_t *cpu_map, /* Set up physical groups */ for (i = 0; i < nr_node_ids; i++) { - SCHED_CPUMASK_VAR(nodemask, allmasks); - SCHED_CPUMASK_VAR(send_covered, allmasks); - + /* FIXME: Use cpumask_of_node */ *nodemask = node_to_cpumask(i); cpus_and(*nodemask, *nodemask, *cpu_map); - if (cpus_empty(*nodemask)) + if (cpumask_empty(nodemask)) continue; init_sched_build_groups(nodemask, cpu_map, @@ -7555,8 +7538,6 @@ static int __build_sched_domains(const cpumask_t *cpu_map, #ifdef CONFIG_NUMA /* Set up node groups */ if (sd_allnodes) { - SCHED_CPUMASK_VAR(send_covered, allmasks); - init_sched_build_groups(cpu_map, cpu_map, &cpu_to_allnodes_group, send_covered, tmpmask); @@ -7565,58 +7546,58 @@ static int __build_sched_domains(const cpumask_t *cpu_map, for (i = 0; i < nr_node_ids; i++) { /* Set up node groups */ struct sched_group *sg, *prev; - SCHED_CPUMASK_VAR(nodemask, allmasks); - SCHED_CPUMASK_VAR(domainspan, allmasks); - SCHED_CPUMASK_VAR(covered, allmasks); int j; + /* FIXME: Use cpumask_of_node */ *nodemask = node_to_cpumask(i); - cpus_clear(*covered); + cpumask_clear(covered); cpus_and(*nodemask, *nodemask, *cpu_map); - if (cpus_empty(*nodemask)) { + if (cpumask_empty(nodemask)) { sched_group_nodes[i] = NULL; continue; } sched_domain_node_span(i, domainspan); - cpus_and(*domainspan, *domainspan, *cpu_map); + cpumask_and(domainspan, domainspan, cpu_map); - sg = kmalloc_node(sizeof(struct sched_group), GFP_KERNEL, i); + sg = kmalloc_node(sizeof(struct sched_group) + cpumask_size(), + GFP_KERNEL, i); if (!sg) { printk(KERN_WARNING "Can not alloc domain group for " "node %d\n", i); goto error; } sched_group_nodes[i] = sg; - for_each_cpu_mask_nr(j, *nodemask) { + for_each_cpu(j, nodemask) { struct sched_domain *sd; sd = &per_cpu(node_domains, j); sd->groups = sg; } sg->__cpu_power = 0; - sg->cpumask = *nodemask; + cpumask_copy(sched_group_cpus(sg), nodemask); sg->next = sg; - cpus_or(*covered, *covered, *nodemask); + cpumask_or(covered, covered, nodemask); prev = sg; for (j = 0; j < nr_node_ids; j++) { - SCHED_CPUMASK_VAR(notcovered, allmasks); int n = (i + j) % nr_node_ids; + /* FIXME: Use cpumask_of_node */ node_to_cpumask_ptr(pnodemask, n); - cpus_complement(*notcovered, *covered); - cpus_and(*tmpmask, *notcovered, *cpu_map); - cpus_and(*tmpmask, *tmpmask, *domainspan); - if (cpus_empty(*tmpmask)) + cpumask_complement(notcovered, covered); + cpumask_and(tmpmask, notcovered, cpu_map); + cpumask_and(tmpmask, tmpmask, domainspan); + if (cpumask_empty(tmpmask)) break; - cpus_and(*tmpmask, *tmpmask, *pnodemask); - if (cpus_empty(*tmpmask)) + cpumask_and(tmpmask, tmpmask, pnodemask); + if (cpumask_empty(tmpmask)) continue; - sg = kmalloc_node(sizeof(struct sched_group), + sg = kmalloc_node(sizeof(struct sched_group) + + cpumask_size(), GFP_KERNEL, i); if (!sg) { printk(KERN_WARNING @@ -7624,9 +7605,9 @@ static int __build_sched_domains(const cpumask_t *cpu_map, goto error; } sg->__cpu_power = 0; - sg->cpumask = *tmpmask; + cpumask_copy(sched_group_cpus(sg), tmpmask); sg->next = prev->next; - cpus_or(*covered, *covered, *tmpmask); + cpumask_or(covered, covered, tmpmask); prev->next = sg; prev = sg; } @@ -7635,22 +7616,22 @@ static int __build_sched_domains(const cpumask_t *cpu_map, /* Calculate CPU power for physical packages and nodes */ #ifdef CONFIG_SCHED_SMT - for_each_cpu_mask_nr(i, *cpu_map) { - struct sched_domain *sd = &per_cpu(cpu_domains, i); + for_each_cpu(i, cpu_map) { + struct sched_domain *sd = &per_cpu(cpu_domains, i).sd; init_sched_groups_power(i, sd); } #endif #ifdef CONFIG_SCHED_MC - for_each_cpu_mask_nr(i, *cpu_map) { - struct sched_domain *sd = &per_cpu(core_domains, i); + for_each_cpu(i, cpu_map) { + struct sched_domain *sd = &per_cpu(core_domains, i).sd; init_sched_groups_power(i, sd); } #endif - for_each_cpu_mask_nr(i, *cpu_map) { - struct sched_domain *sd = &per_cpu(phys_domains, i); + for_each_cpu(i, cpu_map) { + struct sched_domain *sd = &per_cpu(phys_domains, i).sd; init_sched_groups_power(i, sd); } @@ -7662,53 +7643,78 @@ static int __build_sched_domains(const cpumask_t *cpu_map, if (sd_allnodes) { struct sched_group *sg; - cpu_to_allnodes_group(first_cpu(*cpu_map), cpu_map, &sg, + cpu_to_allnodes_group(cpumask_first(cpu_map), cpu_map, &sg, tmpmask); init_numa_sched_groups_power(sg); } #endif /* Attach the domains */ - for_each_cpu_mask_nr(i, *cpu_map) { + for_each_cpu(i, cpu_map) { struct sched_domain *sd; #ifdef CONFIG_SCHED_SMT - sd = &per_cpu(cpu_domains, i); + sd = &per_cpu(cpu_domains, i).sd; #elif defined(CONFIG_SCHED_MC) - sd = &per_cpu(core_domains, i); + sd = &per_cpu(core_domains, i).sd; #else - sd = &per_cpu(phys_domains, i); + sd = &per_cpu(phys_domains, i).sd; #endif cpu_attach_domain(sd, rd, i); } - SCHED_CPUMASK_FREE((void *)allmasks); - return 0; + err = 0; + +free_tmpmask: + free_cpumask_var(tmpmask); +free_send_covered: + free_cpumask_var(send_covered); +free_this_core_map: + free_cpumask_var(this_core_map); +free_this_sibling_map: + free_cpumask_var(this_sibling_map); +free_nodemask: + free_cpumask_var(nodemask); +free_notcovered: +#ifdef CONFIG_NUMA + free_cpumask_var(notcovered); +free_covered: + free_cpumask_var(covered); +free_domainspan: + free_cpumask_var(domainspan); +out: +#endif + return err; + +free_sched_groups: +#ifdef CONFIG_NUMA + kfree(sched_group_nodes); +#endif + goto free_tmpmask; #ifdef CONFIG_NUMA error: free_sched_groups(cpu_map, tmpmask); - SCHED_CPUMASK_FREE((void *)allmasks); - kfree(rd); - return -ENOMEM; + free_rootdomain(rd); + goto free_tmpmask; #endif } -static int build_sched_domains(const cpumask_t *cpu_map) +static int build_sched_domains(const struct cpumask *cpu_map) { return __build_sched_domains(cpu_map, NULL); } -static cpumask_t *doms_cur; /* current sched domains */ +static struct cpumask *doms_cur; /* current sched domains */ static int ndoms_cur; /* number of sched domains in 'doms_cur' */ static struct sched_domain_attr *dattr_cur; /* attribues of custom domains in 'doms_cur' */ /* * Special case: If a kmalloc of a doms_cur partition (array of - * cpumask_t) fails, then fallback to a single sched domain, - * as determined by the single cpumask_t fallback_doms. + * cpumask) fails, then fallback to a single sched domain, + * as determined by the single cpumask fallback_doms. */ -static cpumask_t fallback_doms; +static cpumask_var_t fallback_doms; void __attribute__((weak)) arch_update_cpu_topology(void) { @@ -7719,16 +7725,16 @@ void __attribute__((weak)) arch_update_cpu_topology(void) * For now this just excludes isolated cpus, but could be used to * exclude other special cases in the future. */ -static int arch_init_sched_domains(const cpumask_t *cpu_map) +static int arch_init_sched_domains(const struct cpumask *cpu_map) { int err; arch_update_cpu_topology(); ndoms_cur = 1; - doms_cur = kmalloc(sizeof(cpumask_t), GFP_KERNEL); + doms_cur = kmalloc(cpumask_size(), GFP_KERNEL); if (!doms_cur) - doms_cur = &fallback_doms; - cpus_andnot(*doms_cur, *cpu_map, cpu_isolated_map); + doms_cur = fallback_doms; + cpumask_andnot(doms_cur, cpu_map, cpu_isolated_map); dattr_cur = NULL; err = build_sched_domains(doms_cur); register_sched_domain_sysctl(); @@ -7736,8 +7742,8 @@ static int arch_init_sched_domains(const cpumask_t *cpu_map) return err; } -static void arch_destroy_sched_domains(const cpumask_t *cpu_map, - cpumask_t *tmpmask) +static void arch_destroy_sched_domains(const struct cpumask *cpu_map, + struct cpumask *tmpmask) { free_sched_groups(cpu_map, tmpmask); } @@ -7746,17 +7752,16 @@ static void arch_destroy_sched_domains(const cpumask_t *cpu_map, * Detach sched domains from a group of cpus specified in cpu_map * These cpus will now be attached to the NULL domain */ -static void detach_destroy_domains(const cpumask_t *cpu_map) +static void detach_destroy_domains(const struct cpumask *cpu_map) { - cpumask_t tmpmask; + /* Save because hotplug lock held. */ + static DECLARE_BITMAP(tmpmask, CONFIG_NR_CPUS); int i; - unregister_sched_domain_sysctl(); - - for_each_cpu_mask_nr(i, *cpu_map) + for_each_cpu(i, cpu_map) cpu_attach_domain(NULL, &def_root_domain, i); synchronize_sched(); - arch_destroy_sched_domains(cpu_map, &tmpmask); + arch_destroy_sched_domains(cpu_map, to_cpumask(tmpmask)); } /* handle null as "default" */ @@ -7781,7 +7786,7 @@ static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur, * doms_new[] to the current sched domain partitioning, doms_cur[]. * It destroys each deleted domain and builds each new domain. * - * 'doms_new' is an array of cpumask_t's of length 'ndoms_new'. + * 'doms_new' is an array of cpumask's of length 'ndoms_new'. * The masks don't intersect (don't overlap.) We should setup one * sched domain for each mask. CPUs not in any of the cpumasks will * not be load balanced. If the same cpumask appears both in the @@ -7795,13 +7800,14 @@ static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur, * the single partition 'fallback_doms', it also forces the domains * to be rebuilt. * - * If doms_new == NULL it will be replaced with cpu_online_map. + * If doms_new == NULL it will be replaced with cpu_online_mask. * ndoms_new == 0 is a special case for destroying existing domains, * and it will not create the default domain. * * Call with hotplug lock held */ -void partition_sched_domains(int ndoms_new, cpumask_t *doms_new, +/* FIXME: Change to struct cpumask *doms_new[] */ +void partition_sched_domains(int ndoms_new, struct cpumask *doms_new, struct sched_domain_attr *dattr_new) { int i, j, n; @@ -7816,7 +7822,7 @@ void partition_sched_domains(int ndoms_new, cpumask_t *doms_new, /* Destroy deleted domains */ for (i = 0; i < ndoms_cur; i++) { for (j = 0; j < n; j++) { - if (cpus_equal(doms_cur[i], doms_new[j]) + if (cpumask_equal(&doms_cur[i], &doms_new[j]) && dattrs_equal(dattr_cur, i, dattr_new, j)) goto match1; } @@ -7828,15 +7834,15 @@ match1: if (doms_new == NULL) { ndoms_cur = 0; - doms_new = &fallback_doms; - cpus_andnot(doms_new[0], cpu_online_map, cpu_isolated_map); - dattr_new = NULL; + doms_new = fallback_doms; + cpumask_andnot(&doms_new[0], cpu_online_mask, cpu_isolated_map); + WARN_ON_ONCE(dattr_new); } /* Build new domains */ for (i = 0; i < ndoms_new; i++) { for (j = 0; j < ndoms_cur; j++) { - if (cpus_equal(doms_new[i], doms_cur[j]) + if (cpumask_equal(&doms_new[i], &doms_cur[j]) && dattrs_equal(dattr_new, i, dattr_cur, j)) goto match2; } @@ -7848,7 +7854,7 @@ match2: } /* Remember the new sched domains */ - if (doms_cur != &fallback_doms) + if (doms_cur != fallback_doms) kfree(doms_cur); kfree(dattr_cur); /* kfree(NULL) is safe */ doms_cur = doms_new; @@ -7988,7 +7994,9 @@ static int update_runtime(struct notifier_block *nfb, void __init sched_init_smp(void) { - cpumask_t non_isolated_cpus; + cpumask_var_t non_isolated_cpus; + + alloc_cpumask_var(&non_isolated_cpus, GFP_KERNEL); #if defined(CONFIG_NUMA) sched_group_nodes_bycpu = kzalloc(nr_cpu_ids * sizeof(void **), @@ -7997,10 +8005,10 @@ void __init sched_init_smp(void) #endif get_online_cpus(); mutex_lock(&sched_domains_mutex); - arch_init_sched_domains(&cpu_online_map); - cpus_andnot(non_isolated_cpus, cpu_possible_map, cpu_isolated_map); - if (cpus_empty(non_isolated_cpus)) - cpu_set(smp_processor_id(), non_isolated_cpus); + arch_init_sched_domains(cpu_online_mask); + cpumask_andnot(non_isolated_cpus, cpu_possible_mask, cpu_isolated_map); + if (cpumask_empty(non_isolated_cpus)) + cpumask_set_cpu(smp_processor_id(), non_isolated_cpus); mutex_unlock(&sched_domains_mutex); put_online_cpus(); @@ -8015,9 +8023,13 @@ void __init sched_init_smp(void) init_hrtick(); /* Move init over to a non-isolated CPU */ - if (set_cpus_allowed_ptr(current, &non_isolated_cpus) < 0) + if (set_cpus_allowed_ptr(current, non_isolated_cpus) < 0) BUG(); sched_init_granularity(); + free_cpumask_var(non_isolated_cpus); + + alloc_cpumask_var(&fallback_doms, GFP_KERNEL); + init_sched_rt_class(); } #else void __init sched_init_smp(void) @@ -8332,6 +8344,15 @@ void __init sched_init(void) */ current->sched_class = &fair_sched_class; + /* Allocate the nohz_cpu_mask if CONFIG_CPUMASK_OFFSTACK */ + alloc_bootmem_cpumask_var(&nohz_cpu_mask); +#ifdef CONFIG_SMP +#ifdef CONFIG_NO_HZ + alloc_bootmem_cpumask_var(&nohz.cpu_mask); +#endif + alloc_bootmem_cpumask_var(&cpu_isolated_map); +#endif /* SMP */ + scheduler_running = 1; } @@ -8490,7 +8511,7 @@ static int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent) { struct cfs_rq *cfs_rq; - struct sched_entity *se, *parent_se; + struct sched_entity *se; struct rq *rq; int i; @@ -8506,18 +8527,17 @@ int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent) for_each_possible_cpu(i) { rq = cpu_rq(i); - cfs_rq = kmalloc_node(sizeof(struct cfs_rq), - GFP_KERNEL|__GFP_ZERO, cpu_to_node(i)); + cfs_rq = kzalloc_node(sizeof(struct cfs_rq), + GFP_KERNEL, cpu_to_node(i)); if (!cfs_rq) goto err; - se = kmalloc_node(sizeof(struct sched_entity), - GFP_KERNEL|__GFP_ZERO, cpu_to_node(i)); + se = kzalloc_node(sizeof(struct sched_entity), + GFP_KERNEL, cpu_to_node(i)); if (!se) goto err; - parent_se = parent ? parent->se[i] : NULL; - init_tg_cfs_entry(tg, cfs_rq, se, i, 0, parent_se); + init_tg_cfs_entry(tg, cfs_rq, se, i, 0, parent->se[i]); } return 1; @@ -8578,7 +8598,7 @@ static int alloc_rt_sched_group(struct task_group *tg, struct task_group *parent) { struct rt_rq *rt_rq; - struct sched_rt_entity *rt_se, *parent_se; + struct sched_rt_entity *rt_se; struct rq *rq; int i; @@ -8595,18 +8615,17 @@ int alloc_rt_sched_group(struct task_group *tg, struct task_group *parent) for_each_possible_cpu(i) { rq = cpu_rq(i); - rt_rq = kmalloc_node(sizeof(struct rt_rq), - GFP_KERNEL|__GFP_ZERO, cpu_to_node(i)); + rt_rq = kzalloc_node(sizeof(struct rt_rq), + GFP_KERNEL, cpu_to_node(i)); if (!rt_rq) goto err; - rt_se = kmalloc_node(sizeof(struct sched_rt_entity), - GFP_KERNEL|__GFP_ZERO, cpu_to_node(i)); + rt_se = kzalloc_node(sizeof(struct sched_rt_entity), + GFP_KERNEL, cpu_to_node(i)); if (!rt_se) goto err; - parent_se = parent ? parent->rt_se[i] : NULL; - init_tg_rt_entry(tg, rt_rq, rt_se, i, 0, parent_se); + init_tg_rt_entry(tg, rt_rq, rt_se, i, 0, parent->rt_se[i]); } return 1; @@ -9249,11 +9268,12 @@ struct cgroup_subsys cpu_cgroup_subsys = { * (balbir@in.ibm.com). */ -/* track cpu usage of a group of tasks */ +/* track cpu usage of a group of tasks and its child groups */ struct cpuacct { struct cgroup_subsys_state css; /* cpuusage holds pointer to a u64-type object on every cpu */ u64 *cpuusage; + struct cpuacct *parent; }; struct cgroup_subsys cpuacct_subsys; @@ -9287,6 +9307,9 @@ static struct cgroup_subsys_state *cpuacct_create( return ERR_PTR(-ENOMEM); } + if (cgrp->parent) + ca->parent = cgroup_ca(cgrp->parent); + return &ca->css; } @@ -9366,14 +9389,16 @@ static int cpuacct_populate(struct cgroup_subsys *ss, struct cgroup *cgrp) static void cpuacct_charge(struct task_struct *tsk, u64 cputime) { struct cpuacct *ca; + int cpu; if (!cpuacct_subsys.active) return; + cpu = task_cpu(tsk); ca = task_ca(tsk); - if (ca) { - u64 *cpuusage = percpu_ptr(ca->cpuusage, task_cpu(tsk)); + for (; ca; ca = ca->parent) { + u64 *cpuusage = percpu_ptr(ca->cpuusage, cpu); *cpuusage += cputime; } } diff --git a/kernel/sched_cpupri.c b/kernel/sched_cpupri.c index 52154fefab7..018b7be1db2 100644 --- a/kernel/sched_cpupri.c +++ b/kernel/sched_cpupri.c @@ -67,24 +67,21 @@ static int convert_prio(int prio) * Returns: (int)bool - CPUs were found */ int cpupri_find(struct cpupri *cp, struct task_struct *p, - cpumask_t *lowest_mask) + struct cpumask *lowest_mask) { int idx = 0; int task_pri = convert_prio(p->prio); for_each_cpupri_active(cp->pri_active, idx) { struct cpupri_vec *vec = &cp->pri_to_cpu[idx]; - cpumask_t mask; if (idx >= task_pri) break; - cpus_and(mask, p->cpus_allowed, vec->mask); - - if (cpus_empty(mask)) + if (cpumask_any_and(&p->cpus_allowed, vec->mask) >= nr_cpu_ids) continue; - *lowest_mask = mask; + cpumask_and(lowest_mask, &p->cpus_allowed, vec->mask); return 1; } @@ -126,7 +123,7 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) vec->count--; if (!vec->count) clear_bit(oldpri, cp->pri_active); - cpu_clear(cpu, vec->mask); + cpumask_clear_cpu(cpu, vec->mask); spin_unlock_irqrestore(&vec->lock, flags); } @@ -136,7 +133,7 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) spin_lock_irqsave(&vec->lock, flags); - cpu_set(cpu, vec->mask); + cpumask_set_cpu(cpu, vec->mask); vec->count++; if (vec->count == 1) set_bit(newpri, cp->pri_active); @@ -150,10 +147,11 @@ void cpupri_set(struct cpupri *cp, int cpu, int newpri) /** * cpupri_init - initialize the cpupri structure * @cp: The cpupri context + * @bootmem: true if allocations need to use bootmem * - * Returns: (void) + * Returns: -ENOMEM if memory fails. */ -void cpupri_init(struct cpupri *cp) +int cpupri_init(struct cpupri *cp, bool bootmem) { int i; @@ -164,11 +162,30 @@ void cpupri_init(struct cpupri *cp) spin_lock_init(&vec->lock); vec->count = 0; - cpus_clear(vec->mask); + if (bootmem) + alloc_bootmem_cpumask_var(&vec->mask); + else if (!alloc_cpumask_var(&vec->mask, GFP_KERNEL)) + goto cleanup; } for_each_possible_cpu(i) cp->cpu_to_pri[i] = CPUPRI_INVALID; + return 0; + +cleanup: + for (i--; i >= 0; i--) + free_cpumask_var(cp->pri_to_cpu[i].mask); + return -ENOMEM; } +/** + * cpupri_cleanup - clean up the cpupri structure + * @cp: The cpupri context + */ +void cpupri_cleanup(struct cpupri *cp) +{ + int i; + for (i = 0; i < CPUPRI_NR_PRIORITIES; i++) + free_cpumask_var(cp->pri_to_cpu[i].mask); +} diff --git a/kernel/sched_cpupri.h b/kernel/sched_cpupri.h index f25811b0f93..642a94ef8a0 100644 --- a/kernel/sched_cpupri.h +++ b/kernel/sched_cpupri.h @@ -14,7 +14,7 @@ struct cpupri_vec { spinlock_t lock; int count; - cpumask_t mask; + cpumask_var_t mask; }; struct cpupri { @@ -27,7 +27,8 @@ struct cpupri { int cpupri_find(struct cpupri *cp, struct task_struct *p, cpumask_t *lowest_mask); void cpupri_set(struct cpupri *cp, int cpu, int pri); -void cpupri_init(struct cpupri *cp); +int cpupri_init(struct cpupri *cp, bool bootmem); +void cpupri_cleanup(struct cpupri *cp); #else #define cpupri_set(cp, cpu, pri) do { } while (0) #define cpupri_init() do { } while (0) diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 26ed8e3d1c1..baf2f17af46 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -53,6 +53,40 @@ static unsigned long nsec_low(unsigned long long nsec) #define SPLIT_NS(x) nsec_high(x), nsec_low(x) +#ifdef CONFIG_FAIR_GROUP_SCHED +static void print_cfs_group_stats(struct seq_file *m, int cpu, + struct task_group *tg) +{ + struct sched_entity *se = tg->se[cpu]; + if (!se) + return; + +#define P(F) \ + SEQ_printf(m, " .%-30s: %lld\n", #F, (long long)F) +#define PN(F) \ + SEQ_printf(m, " .%-30s: %lld.%06ld\n", #F, SPLIT_NS((long long)F)) + + PN(se->exec_start); + PN(se->vruntime); + PN(se->sum_exec_runtime); +#ifdef CONFIG_SCHEDSTATS + PN(se->wait_start); + PN(se->sleep_start); + PN(se->block_start); + PN(se->sleep_max); + PN(se->block_max); + PN(se->exec_max); + PN(se->slice_max); + PN(se->wait_max); + PN(se->wait_sum); + P(se->wait_count); +#endif + P(se->load.weight); +#undef PN +#undef P +} +#endif + static void print_task(struct seq_file *m, struct rq *rq, struct task_struct *p) { @@ -121,14 +155,9 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) #if defined(CONFIG_CGROUP_SCHED) && defined(CONFIG_FAIR_GROUP_SCHED) char path[128] = ""; - struct cgroup *cgroup = NULL; struct task_group *tg = cfs_rq->tg; - if (tg) - cgroup = tg->css.cgroup; - - if (cgroup) - cgroup_path(cgroup, path, sizeof(path)); + cgroup_path(tg->css.cgroup, path, sizeof(path)); SEQ_printf(m, "\ncfs_rq[%d]:%s\n", cpu, path); #else @@ -168,6 +197,7 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) #ifdef CONFIG_SMP SEQ_printf(m, " .%-30s: %lu\n", "shares", cfs_rq->shares); #endif + print_cfs_group_stats(m, cpu, cfs_rq->tg); #endif } @@ -175,14 +205,9 @@ void print_rt_rq(struct seq_file *m, int cpu, struct rt_rq *rt_rq) { #if defined(CONFIG_CGROUP_SCHED) && defined(CONFIG_RT_GROUP_SCHED) char path[128] = ""; - struct cgroup *cgroup = NULL; struct task_group *tg = rt_rq->tg; - if (tg) - cgroup = tg->css.cgroup; - - if (cgroup) - cgroup_path(cgroup, path, sizeof(path)); + cgroup_path(tg->css.cgroup, path, sizeof(path)); SEQ_printf(m, "\nrt_rq[%d]:%s\n", cpu, path); #else @@ -272,7 +297,7 @@ static int sched_debug_show(struct seq_file *m, void *v) u64 now = ktime_to_ns(ktime_get()); int cpu; - SEQ_printf(m, "Sched Debug Version: v0.07, %s %.*s\n", + SEQ_printf(m, "Sched Debug Version: v0.08, %s %.*s\n", init_utsname()->release, (int)strcspn(init_utsname()->version, " "), init_utsname()->version); diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index 98345e45b05..08ffffd4a41 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -1017,14 +1017,13 @@ static void yield_task_fair(struct rq *rq) * search starts with cpus closest then further out as needed, * so we always favor a closer, idle cpu. * Domains may include CPUs that are not usable for migration, - * hence we need to mask them out (cpu_active_map) + * hence we need to mask them out (cpu_active_mask) * * Returns the CPU we should wake onto. */ #if defined(ARCH_HAS_SCHED_WAKE_IDLE) static int wake_idle(int cpu, struct task_struct *p) { - cpumask_t tmp; struct sched_domain *sd; int i; @@ -1044,10 +1043,9 @@ static int wake_idle(int cpu, struct task_struct *p) if ((sd->flags & SD_WAKE_IDLE) || ((sd->flags & SD_WAKE_IDLE_FAR) && !task_hot(p, task_rq(p)->clock, sd))) { - cpus_and(tmp, sd->span, p->cpus_allowed); - cpus_and(tmp, tmp, cpu_active_map); - for_each_cpu_mask_nr(i, tmp) { - if (idle_cpu(i)) { + for_each_cpu_and(i, sched_domain_span(sd), + &p->cpus_allowed) { + if (cpu_active(i) && idle_cpu(i)) { if (i != task_cpu(p)) { schedstat_inc(p, se.nr_wakeups_idle); @@ -1240,13 +1238,13 @@ static int select_task_rq_fair(struct task_struct *p, int sync) * this_cpu and prev_cpu are present in: */ for_each_domain(this_cpu, sd) { - if (cpu_isset(prev_cpu, sd->span)) { + if (cpumask_test_cpu(prev_cpu, sched_domain_span(sd))) { this_sd = sd; break; } } - if (unlikely(!cpu_isset(this_cpu, p->cpus_allowed))) + if (unlikely(!cpumask_test_cpu(this_cpu, &p->cpus_allowed))) goto out; /* diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index d9ba9d5f99d..94aab72f6a0 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -15,7 +15,7 @@ static inline void rt_set_overload(struct rq *rq) if (!rq->online) return; - cpu_set(rq->cpu, rq->rd->rto_mask); + cpumask_set_cpu(rq->cpu, rq->rd->rto_mask); /* * Make sure the mask is visible before we set * the overload count. That is checked to determine @@ -34,7 +34,7 @@ static inline void rt_clear_overload(struct rq *rq) /* the order here really doesn't matter */ atomic_dec(&rq->rd->rto_count); - cpu_clear(rq->cpu, rq->rd->rto_mask); + cpumask_clear_cpu(rq->cpu, rq->rd->rto_mask); } static void update_rt_migration(struct rq *rq) @@ -139,14 +139,14 @@ static int rt_se_boosted(struct sched_rt_entity *rt_se) } #ifdef CONFIG_SMP -static inline cpumask_t sched_rt_period_mask(void) +static inline const struct cpumask *sched_rt_period_mask(void) { return cpu_rq(smp_processor_id())->rd->span; } #else -static inline cpumask_t sched_rt_period_mask(void) +static inline const struct cpumask *sched_rt_period_mask(void) { - return cpu_online_map; + return cpu_online_mask; } #endif @@ -212,9 +212,9 @@ static inline int rt_rq_throttled(struct rt_rq *rt_rq) return rt_rq->rt_throttled; } -static inline cpumask_t sched_rt_period_mask(void) +static inline const struct cpumask *sched_rt_period_mask(void) { - return cpu_online_map; + return cpu_online_mask; } static inline @@ -241,11 +241,11 @@ static int do_balance_runtime(struct rt_rq *rt_rq) int i, weight, more = 0; u64 rt_period; - weight = cpus_weight(rd->span); + weight = cpumask_weight(rd->span); spin_lock(&rt_b->rt_runtime_lock); rt_period = ktime_to_ns(rt_b->rt_period); - for_each_cpu_mask_nr(i, rd->span) { + for_each_cpu(i, rd->span) { struct rt_rq *iter = sched_rt_period_rt_rq(rt_b, i); s64 diff; @@ -324,7 +324,7 @@ static void __disable_runtime(struct rq *rq) /* * Greedy reclaim, take back as much as we can. */ - for_each_cpu_mask(i, rd->span) { + for_each_cpu(i, rd->span) { struct rt_rq *iter = sched_rt_period_rt_rq(rt_b, i); s64 diff; @@ -429,13 +429,13 @@ static inline int balance_runtime(struct rt_rq *rt_rq) static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun) { int i, idle = 1; - cpumask_t span; + const struct cpumask *span; if (!rt_bandwidth_enabled() || rt_b->rt_runtime == RUNTIME_INF) return 1; span = sched_rt_period_mask(); - for_each_cpu_mask(i, span) { + for_each_cpu(i, span) { int enqueue = 0; struct rt_rq *rt_rq = sched_rt_period_rt_rq(rt_b, i); struct rq *rq = rq_of_rt_rq(rt_rq); @@ -537,13 +537,13 @@ static void update_curr_rt(struct rq *rq) for_each_sched_rt_entity(rt_se) { rt_rq = rt_rq_of_se(rt_se); - spin_lock(&rt_rq->rt_runtime_lock); if (sched_rt_runtime(rt_rq) != RUNTIME_INF) { + spin_lock(&rt_rq->rt_runtime_lock); rt_rq->rt_time += delta_exec; if (sched_rt_runtime_exceeded(rt_rq)) resched_task(curr); + spin_unlock(&rt_rq->rt_runtime_lock); } - spin_unlock(&rt_rq->rt_runtime_lock); } } @@ -805,17 +805,20 @@ static int select_task_rq_rt(struct task_struct *p, int sync) static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p) { - cpumask_t mask; + cpumask_var_t mask; if (rq->curr->rt.nr_cpus_allowed == 1) return; - if (p->rt.nr_cpus_allowed != 1 - && cpupri_find(&rq->rd->cpupri, p, &mask)) + if (!alloc_cpumask_var(&mask, GFP_ATOMIC)) return; - if (!cpupri_find(&rq->rd->cpupri, rq->curr, &mask)) - return; + if (p->rt.nr_cpus_allowed != 1 + && cpupri_find(&rq->rd->cpupri, p, mask)) + goto free; + + if (!cpupri_find(&rq->rd->cpupri, rq->curr, mask)) + goto free; /* * There appears to be other cpus that can accept @@ -824,6 +827,8 @@ static void check_preempt_equal_prio(struct rq *rq, struct task_struct *p) */ requeue_task_rt(rq, p, 1); resched_task(rq->curr); +free: + free_cpumask_var(mask); } #endif /* CONFIG_SMP */ @@ -910,14 +915,15 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p) #define RT_MAX_TRIES 3 static int double_lock_balance(struct rq *this_rq, struct rq *busiest); -static void double_unlock_balance(struct rq *this_rq, struct rq *busiest); +static inline void double_unlock_balance(struct rq *this_rq, + struct rq *busiest); static void deactivate_task(struct rq *rq, struct task_struct *p, int sleep); static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu) { if (!task_running(rq, p) && - (cpu < 0 || cpu_isset(cpu, p->cpus_allowed)) && + (cpu < 0 || cpumask_test_cpu(cpu, &p->cpus_allowed)) && (p->rt.nr_cpus_allowed > 1)) return 1; return 0; @@ -956,7 +962,7 @@ static struct task_struct *pick_next_highest_task_rt(struct rq *rq, int cpu) return next; } -static DEFINE_PER_CPU(cpumask_t, local_cpu_mask); +static DEFINE_PER_CPU(cpumask_var_t, local_cpu_mask); static inline int pick_optimal_cpu(int this_cpu, cpumask_t *mask) { @@ -976,7 +982,7 @@ static inline int pick_optimal_cpu(int this_cpu, cpumask_t *mask) static int find_lowest_rq(struct task_struct *task) { struct sched_domain *sd; - cpumask_t *lowest_mask = &__get_cpu_var(local_cpu_mask); + struct cpumask *lowest_mask = __get_cpu_var(local_cpu_mask); int this_cpu = smp_processor_id(); int cpu = task_cpu(task); @@ -991,7 +997,7 @@ static int find_lowest_rq(struct task_struct *task) * I guess we might want to change cpupri_find() to ignore those * in the first place. */ - cpus_and(*lowest_mask, *lowest_mask, cpu_active_map); + cpumask_and(lowest_mask, lowest_mask, cpu_active_mask); /* * At this point we have built a mask of cpus representing the @@ -1001,7 +1007,7 @@ static int find_lowest_rq(struct task_struct *task) * We prioritize the last cpu that the task executed on since * it is most likely cache-hot in that location. */ - if (cpu_isset(cpu, *lowest_mask)) + if (cpumask_test_cpu(cpu, lowest_mask)) return cpu; /* @@ -1016,7 +1022,8 @@ static int find_lowest_rq(struct task_struct *task) cpumask_t domain_mask; int best_cpu; - cpus_and(domain_mask, sd->span, *lowest_mask); + cpumask_and(&domain_mask, sched_domain_span(sd), + lowest_mask); best_cpu = pick_optimal_cpu(this_cpu, &domain_mask); @@ -1057,8 +1064,8 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq) * Also make sure that it wasn't scheduled on its rq. */ if (unlikely(task_rq(task) != rq || - !cpu_isset(lowest_rq->cpu, - task->cpus_allowed) || + !cpumask_test_cpu(lowest_rq->cpu, + &task->cpus_allowed) || task_running(rq, task) || !task->se.on_rq)) { @@ -1179,7 +1186,7 @@ static int pull_rt_task(struct rq *this_rq) next = pick_next_task_rt(this_rq); - for_each_cpu_mask_nr(cpu, this_rq->rd->rto_mask) { + for_each_cpu(cpu, this_rq->rd->rto_mask) { if (this_cpu == cpu) continue; @@ -1308,9 +1315,9 @@ move_one_task_rt(struct rq *this_rq, int this_cpu, struct rq *busiest, } static void set_cpus_allowed_rt(struct task_struct *p, - const cpumask_t *new_mask) + const struct cpumask *new_mask) { - int weight = cpus_weight(*new_mask); + int weight = cpumask_weight(new_mask); BUG_ON(!rt_task(p)); @@ -1331,7 +1338,7 @@ static void set_cpus_allowed_rt(struct task_struct *p, update_rt_migration(rq); } - p->cpus_allowed = *new_mask; + cpumask_copy(&p->cpus_allowed, new_mask); p->rt.nr_cpus_allowed = weight; } @@ -1374,6 +1381,14 @@ static void switched_from_rt(struct rq *rq, struct task_struct *p, if (!rq->rt.rt_nr_running) pull_rt_task(rq); } + +static inline void init_sched_rt_class(void) +{ + unsigned int i; + + for_each_possible_cpu(i) + alloc_cpumask_var(&per_cpu(local_cpu_mask, i), GFP_KERNEL); +} #endif /* CONFIG_SMP */ /* @@ -1544,3 +1559,4 @@ static void print_rt_stats(struct seq_file *m, int cpu) rcu_read_unlock(); } #endif /* CONFIG_SCHED_DEBUG */ + diff --git a/kernel/sched_stats.h b/kernel/sched_stats.h index 7dbf72a2b02..ce340835d05 100644 --- a/kernel/sched_stats.h +++ b/kernel/sched_stats.h @@ -42,7 +42,8 @@ static int show_schedstat(struct seq_file *seq, void *v) for_each_domain(cpu, sd) { enum cpu_idle_type itype; - cpumask_scnprintf(mask_str, mask_len, sd->span); + cpumask_scnprintf(mask_str, mask_len, + *sched_domain_span(sd)); seq_printf(seq, "domain%d %s", dcount++, mask_str); for (itype = CPU_IDLE; itype < CPU_MAX_IDLE_TYPES; itype++) { diff --git a/kernel/signal.c b/kernel/signal.c index 4530fc65445..e9afe63da24 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -41,6 +41,8 @@ static struct kmem_cache *sigqueue_cachep; +DEFINE_TRACE(sched_signal_send); + static void __user *sig_handler(struct task_struct *t, int sig) { return t->sighand->action[sig - 1].sa.sa_handler; diff --git a/kernel/softlockup.c b/kernel/softlockup.c index 3953e4aed73..884e6cd2769 100644 --- a/kernel/softlockup.c +++ b/kernel/softlockup.c @@ -164,7 +164,7 @@ unsigned long __read_mostly sysctl_hung_task_check_count = 1024; /* * Zero means infinite timeout - no checking done: */ -unsigned long __read_mostly sysctl_hung_task_timeout_secs = 120; +unsigned long __read_mostly sysctl_hung_task_timeout_secs = 480; unsigned long __read_mostly sysctl_hung_task_warnings = 10; diff --git a/kernel/sys.c b/kernel/sys.c index 31deba8f7d1..5fc3a0cfb99 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -858,8 +858,8 @@ void do_sys_times(struct tms *tms) struct task_cputime cputime; cputime_t cutime, cstime; - spin_lock_irq(¤t->sighand->siglock); thread_group_cputime(current, &cputime); + spin_lock_irq(¤t->sighand->siglock); cutime = current->signal->cutime; cstime = current->signal->cstime; spin_unlock_irq(¤t->sighand->siglock); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 3d56fe7570d..c83f566e940 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -487,6 +487,16 @@ static struct ctl_table kern_table[] = { .proc_handler = &ftrace_enable_sysctl, }, #endif +#ifdef CONFIG_TRACING + { + .ctl_name = CTL_UNNUMBERED, + .procname = "ftrace_dump_on_oops", + .data = &ftrace_dump_on_oops, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, +#endif #ifdef CONFIG_MODULES { .ctl_name = KERN_MODPROBE, diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 342fc9ccab4..70f872c71f4 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -144,7 +144,7 @@ void tick_nohz_update_jiffies(void) if (!ts->tick_stopped) return; - cpu_clear(cpu, nohz_cpu_mask); + cpumask_clear_cpu(cpu, nohz_cpu_mask); now = ktime_get(); ts->idle_waketime = now; @@ -283,7 +283,7 @@ void tick_nohz_stop_sched_tick(int inidle) if ((long)delta_jiffies >= 1) { if (delta_jiffies > 1) - cpu_set(cpu, nohz_cpu_mask); + cpumask_set_cpu(cpu, nohz_cpu_mask); /* * nohz_stop_sched_tick can be called several times before * the nohz_restart_sched_tick is called. This happens when @@ -296,7 +296,7 @@ void tick_nohz_stop_sched_tick(int inidle) /* * sched tick not stopped! */ - cpu_clear(cpu, nohz_cpu_mask); + cpumask_clear_cpu(cpu, nohz_cpu_mask); goto out; } @@ -354,7 +354,7 @@ void tick_nohz_stop_sched_tick(int inidle) * softirq. */ tick_do_update_jiffies64(ktime_get()); - cpu_clear(cpu, nohz_cpu_mask); + cpumask_clear_cpu(cpu, nohz_cpu_mask); } raise_softirq_irqoff(TIMER_SOFTIRQ); out: @@ -432,7 +432,7 @@ void tick_nohz_restart_sched_tick(void) select_nohz_load_balancer(0); now = ktime_get(); tick_do_update_jiffies64(now); - cpu_clear(cpu, nohz_cpu_mask); + cpumask_clear_cpu(cpu, nohz_cpu_mask); /* * We stopped the tick in idle. Update process times would miss the diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 33dbefd471e..9cbf7761f49 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -3,12 +3,25 @@ # select HAVE_FUNCTION_TRACER: # +config USER_STACKTRACE_SUPPORT + bool + config NOP_TRACER bool config HAVE_FUNCTION_TRACER bool +config HAVE_FUNCTION_RET_TRACER + bool + +config HAVE_FUNCTION_TRACE_MCOUNT_TEST + bool + help + This gets selected when the arch tests the function_trace_stop + variable at the mcount call site. Otherwise, this variable + is tested by the called function. + config HAVE_DYNAMIC_FTRACE bool @@ -47,6 +60,16 @@ config FUNCTION_TRACER (the bootup default), then the overhead of the instructions is very small and not measurable even in micro-benchmarks. +config FUNCTION_RET_TRACER + bool "Kernel Function return Tracer" + depends on HAVE_FUNCTION_RET_TRACER + depends on FUNCTION_TRACER + help + Enable the kernel to trace a function at its return. + It's first purpose is to trace the duration of functions. + This is done by setting the current return address on the thread + info structure of the current task. + config IRQSOFF_TRACER bool "Interrupts-off Latency Tracer" default n @@ -138,6 +161,59 @@ config BOOT_TRACER selected, because the self-tests are an initcall as well and that would invalidate the boot trace. ) +config TRACE_BRANCH_PROFILING + bool "Trace likely/unlikely profiler" + depends on DEBUG_KERNEL + select TRACING + help + This tracer profiles all the the likely and unlikely macros + in the kernel. It will display the results in: + + /debugfs/tracing/profile_annotated_branch + + Note: this will add a significant overhead, only turn this + on if you need to profile the system's use of these macros. + + Say N if unsure. + +config PROFILE_ALL_BRANCHES + bool "Profile all if conditionals" + depends on TRACE_BRANCH_PROFILING + help + This tracer profiles all branch conditions. Every if () + taken in the kernel is recorded whether it hit or miss. + The results will be displayed in: + + /debugfs/tracing/profile_branch + + This configuration, when enabled, will impose a great overhead + on the system. This should only be enabled when the system + is to be analyzed + + Say N if unsure. + +config TRACING_BRANCHES + bool + help + Selected by tracers that will trace the likely and unlikely + conditions. This prevents the tracers themselves from being + profiled. Profiling the tracing infrastructure can only happen + when the likelys and unlikelys are not being traced. + +config BRANCH_TRACER + bool "Trace likely/unlikely instances" + depends on TRACE_BRANCH_PROFILING + select TRACING_BRANCHES + help + This traces the events of likely and unlikely condition + calls in the kernel. The difference between this and the + "Trace likely/unlikely profiler" is that this is not a + histogram of the callers, but actually places the calling + events into a running trace buffer to see when and where the + events happened, as well as their results. + + Say N if unsure. + config STACK_TRACER bool "Trace max stack" depends on HAVE_FUNCTION_TRACER diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index c8228b1a49e..1a8c9259dc6 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -10,6 +10,11 @@ CFLAGS_trace_selftest_dynamic.o = -pg obj-y += trace_selftest_dynamic.o endif +# If unlikely tracing is enabled, do not trace these files +ifdef CONFIG_TRACING_BRANCHES +KBUILD_CFLAGS += -DDISABLE_BRANCH_PROFILING +endif + obj-$(CONFIG_FUNCTION_TRACER) += libftrace.o obj-$(CONFIG_RING_BUFFER) += ring_buffer.o @@ -24,5 +29,7 @@ obj-$(CONFIG_NOP_TRACER) += trace_nop.o obj-$(CONFIG_STACK_TRACER) += trace_stack.o obj-$(CONFIG_MMIOTRACE) += trace_mmiotrace.o obj-$(CONFIG_BOOT_TRACER) += trace_boot.o +obj-$(CONFIG_FUNCTION_RET_TRACER) += trace_functions_return.o +obj-$(CONFIG_TRACE_BRANCH_PROFILING) += trace_branch.o libftrace-y := ftrace.o diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 78db083390f..53042f118f2 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -47,6 +47,12 @@ int ftrace_enabled __read_mostly; static int last_ftrace_enabled; +/* Quick disabling of function tracer. */ +int function_trace_stop; + +/* By default, current tracing type is normal tracing. */ +enum ftrace_tracing_type_t ftrace_tracing_type = FTRACE_TYPE_ENTER; + /* * ftrace_disabled is set when an anomaly is discovered. * ftrace_disabled is much stronger than ftrace_enabled. @@ -63,6 +69,7 @@ static struct ftrace_ops ftrace_list_end __read_mostly = static struct ftrace_ops *ftrace_list __read_mostly = &ftrace_list_end; ftrace_func_t ftrace_trace_function __read_mostly = ftrace_stub; +ftrace_func_t __ftrace_trace_function __read_mostly = ftrace_stub; static void ftrace_list_func(unsigned long ip, unsigned long parent_ip) { @@ -88,8 +95,23 @@ static void ftrace_list_func(unsigned long ip, unsigned long parent_ip) void clear_ftrace_function(void) { ftrace_trace_function = ftrace_stub; + __ftrace_trace_function = ftrace_stub; } +#ifndef CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST +/* + * For those archs that do not test ftrace_trace_stop in their + * mcount call site, we need to do it from C. + */ +static void ftrace_test_stop_func(unsigned long ip, unsigned long parent_ip) +{ + if (function_trace_stop) + return; + + __ftrace_trace_function(ip, parent_ip); +} +#endif + static int __register_ftrace_function(struct ftrace_ops *ops) { /* should not be called from interrupt context */ @@ -110,10 +132,18 @@ static int __register_ftrace_function(struct ftrace_ops *ops) * For one func, simply call it directly. * For more than one func, call the chain. */ +#ifdef CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST if (ops->next == &ftrace_list_end) ftrace_trace_function = ops->func; else ftrace_trace_function = ftrace_list_func; +#else + if (ops->next == &ftrace_list_end) + __ftrace_trace_function = ops->func; + else + __ftrace_trace_function = ftrace_list_func; + ftrace_trace_function = ftrace_test_stop_func; +#endif } spin_unlock(&ftrace_lock); @@ -152,8 +182,7 @@ static int __unregister_ftrace_function(struct ftrace_ops *ops) if (ftrace_enabled) { /* If we only have one func left, then call that directly */ - if (ftrace_list == &ftrace_list_end || - ftrace_list->next == &ftrace_list_end) + if (ftrace_list->next == &ftrace_list_end) ftrace_trace_function = ftrace_list->func; } @@ -308,7 +337,7 @@ ftrace_record_ip(unsigned long ip) { struct dyn_ftrace *rec; - if (!ftrace_enabled || ftrace_disabled) + if (ftrace_disabled) return NULL; rec = ftrace_alloc_dyn_node(ip); @@ -322,14 +351,58 @@ ftrace_record_ip(unsigned long ip) return rec; } -#define FTRACE_ADDR ((long)(ftrace_caller)) +static void print_ip_ins(const char *fmt, unsigned char *p) +{ + int i; + + printk(KERN_CONT "%s", fmt); + + for (i = 0; i < MCOUNT_INSN_SIZE; i++) + printk(KERN_CONT "%s%02x", i ? ":" : "", p[i]); +} + +static void ftrace_bug(int failed, unsigned long ip) +{ + switch (failed) { + case -EFAULT: + FTRACE_WARN_ON_ONCE(1); + pr_info("ftrace faulted on modifying "); + print_ip_sym(ip); + break; + case -EINVAL: + FTRACE_WARN_ON_ONCE(1); + pr_info("ftrace failed to modify "); + print_ip_sym(ip); + print_ip_ins(" actual: ", (unsigned char *)ip); + printk(KERN_CONT "\n"); + break; + case -EPERM: + FTRACE_WARN_ON_ONCE(1); + pr_info("ftrace faulted on writing "); + print_ip_sym(ip); + break; + default: + FTRACE_WARN_ON_ONCE(1); + pr_info("ftrace faulted on unknown error "); + print_ip_sym(ip); + } +} + static int -__ftrace_replace_code(struct dyn_ftrace *rec, - unsigned char *nop, int enable) +__ftrace_replace_code(struct dyn_ftrace *rec, int enable) { unsigned long ip, fl; - unsigned char *call, *old, *new; + unsigned long ftrace_addr; + +#ifdef CONFIG_FUNCTION_RET_TRACER + if (ftrace_tracing_type == FTRACE_TYPE_ENTER) + ftrace_addr = (unsigned long)ftrace_caller; + else + ftrace_addr = (unsigned long)ftrace_return_caller; +#else + ftrace_addr = (unsigned long)ftrace_caller; +#endif ip = rec->ip; @@ -388,34 +461,28 @@ __ftrace_replace_code(struct dyn_ftrace *rec, } } - call = ftrace_call_replace(ip, FTRACE_ADDR); - - if (rec->flags & FTRACE_FL_ENABLED) { - old = nop; - new = call; - } else { - old = call; - new = nop; - } - - return ftrace_modify_code(ip, old, new); + if (rec->flags & FTRACE_FL_ENABLED) + return ftrace_make_call(rec, ftrace_addr); + else + return ftrace_make_nop(NULL, rec, ftrace_addr); } static void ftrace_replace_code(int enable) { int i, failed; - unsigned char *nop = NULL; struct dyn_ftrace *rec; struct ftrace_page *pg; - nop = ftrace_nop_replace(); - for (pg = ftrace_pages_start; pg; pg = pg->next) { for (i = 0; i < pg->index; i++) { rec = &pg->records[i]; - /* don't modify code that has already faulted */ - if (rec->flags & FTRACE_FL_FAILED) + /* + * Skip over free records and records that have + * failed. + */ + if (rec->flags & FTRACE_FL_FREE || + rec->flags & FTRACE_FL_FAILED) continue; /* ignore updates to this record's mcount site */ @@ -426,68 +493,30 @@ static void ftrace_replace_code(int enable) unfreeze_record(rec); } - failed = __ftrace_replace_code(rec, nop, enable); + failed = __ftrace_replace_code(rec, enable); if (failed && (rec->flags & FTRACE_FL_CONVERTED)) { rec->flags |= FTRACE_FL_FAILED; if ((system_state == SYSTEM_BOOTING) || !core_kernel_text(rec->ip)) { ftrace_free_rec(rec); - } + } else + ftrace_bug(failed, rec->ip); } } } } -static void print_ip_ins(const char *fmt, unsigned char *p) -{ - int i; - - printk(KERN_CONT "%s", fmt); - - for (i = 0; i < MCOUNT_INSN_SIZE; i++) - printk(KERN_CONT "%s%02x", i ? ":" : "", p[i]); -} - static int -ftrace_code_disable(struct dyn_ftrace *rec) +ftrace_code_disable(struct module *mod, struct dyn_ftrace *rec) { unsigned long ip; - unsigned char *nop, *call; int ret; ip = rec->ip; - nop = ftrace_nop_replace(); - call = ftrace_call_replace(ip, mcount_addr); - - ret = ftrace_modify_code(ip, call, nop); + ret = ftrace_make_nop(mod, rec, mcount_addr); if (ret) { - switch (ret) { - case -EFAULT: - FTRACE_WARN_ON_ONCE(1); - pr_info("ftrace faulted on modifying "); - print_ip_sym(ip); - break; - case -EINVAL: - FTRACE_WARN_ON_ONCE(1); - pr_info("ftrace failed to modify "); - print_ip_sym(ip); - print_ip_ins(" expected: ", call); - print_ip_ins(" actual: ", (unsigned char *)ip); - print_ip_ins(" replace: ", nop); - printk(KERN_CONT "\n"); - break; - case -EPERM: - FTRACE_WARN_ON_ONCE(1); - pr_info("ftrace faulted on writing "); - print_ip_sym(ip); - break; - default: - FTRACE_WARN_ON_ONCE(1); - pr_info("ftrace faulted on unknown error "); - print_ip_sym(ip); - } - + ftrace_bug(ret, ip); rec->flags |= FTRACE_FL_FAILED; return 0; } @@ -515,7 +544,7 @@ static void ftrace_run_update_code(int command) } static ftrace_func_t saved_ftrace_func; -static int ftrace_start; +static int ftrace_start_up; static DEFINE_MUTEX(ftrace_start_lock); static void ftrace_startup(void) @@ -526,7 +555,7 @@ static void ftrace_startup(void) return; mutex_lock(&ftrace_start_lock); - ftrace_start++; + ftrace_start_up++; command |= FTRACE_ENABLE_CALLS; if (saved_ftrace_func != ftrace_trace_function) { @@ -550,8 +579,8 @@ static void ftrace_shutdown(void) return; mutex_lock(&ftrace_start_lock); - ftrace_start--; - if (!ftrace_start) + ftrace_start_up--; + if (!ftrace_start_up) command |= FTRACE_DISABLE_CALLS; if (saved_ftrace_func != ftrace_trace_function) { @@ -577,8 +606,8 @@ static void ftrace_startup_sysctl(void) mutex_lock(&ftrace_start_lock); /* Force update next time */ saved_ftrace_func = NULL; - /* ftrace_start is true if we want ftrace running */ - if (ftrace_start) + /* ftrace_start_up is true if we want ftrace running */ + if (ftrace_start_up) command |= FTRACE_ENABLE_CALLS; ftrace_run_update_code(command); @@ -593,8 +622,8 @@ static void ftrace_shutdown_sysctl(void) return; mutex_lock(&ftrace_start_lock); - /* ftrace_start is true if ftrace is running */ - if (ftrace_start) + /* ftrace_start_up is true if ftrace is running */ + if (ftrace_start_up) command |= FTRACE_DISABLE_CALLS; ftrace_run_update_code(command); @@ -605,7 +634,7 @@ static cycle_t ftrace_update_time; static unsigned long ftrace_update_cnt; unsigned long ftrace_update_tot_cnt; -static int ftrace_update_code(void) +static int ftrace_update_code(struct module *mod) { struct dyn_ftrace *p, *t; cycle_t start, stop; @@ -622,7 +651,7 @@ static int ftrace_update_code(void) list_del_init(&p->list); /* convert record (i.e, patch mcount-call with NOP) */ - if (ftrace_code_disable(p)) { + if (ftrace_code_disable(mod, p)) { p->flags |= FTRACE_FL_CONVERTED; ftrace_update_cnt++; } else @@ -1181,7 +1210,7 @@ ftrace_regex_release(struct inode *inode, struct file *file, int enable) mutex_lock(&ftrace_sysctl_lock); mutex_lock(&ftrace_start_lock); - if (ftrace_start && ftrace_enabled) + if (ftrace_start_up && ftrace_enabled) ftrace_run_update_code(FTRACE_ENABLE_CALLS); mutex_unlock(&ftrace_start_lock); mutex_unlock(&ftrace_sysctl_lock); @@ -1268,7 +1297,8 @@ static __init int ftrace_init_debugfs(void) fs_initcall(ftrace_init_debugfs); -static int ftrace_convert_nops(unsigned long *start, +static int ftrace_convert_nops(struct module *mod, + unsigned long *start, unsigned long *end) { unsigned long *p; @@ -1279,23 +1309,32 @@ static int ftrace_convert_nops(unsigned long *start, p = start; while (p < end) { addr = ftrace_call_adjust(*p++); + /* + * Some architecture linkers will pad between + * the different mcount_loc sections of different + * object files to satisfy alignments. + * Skip any NULL pointers. + */ + if (!addr) + continue; ftrace_record_ip(addr); } /* disable interrupts to prevent kstop machine */ local_irq_save(flags); - ftrace_update_code(); + ftrace_update_code(mod); local_irq_restore(flags); mutex_unlock(&ftrace_start_lock); return 0; } -void ftrace_init_module(unsigned long *start, unsigned long *end) +void ftrace_init_module(struct module *mod, + unsigned long *start, unsigned long *end) { if (ftrace_disabled || start == end) return; - ftrace_convert_nops(start, end); + ftrace_convert_nops(mod, start, end); } extern unsigned long __start_mcount_loc[]; @@ -1325,7 +1364,8 @@ void __init ftrace_init(void) last_ftrace_enabled = ftrace_enabled = 1; - ret = ftrace_convert_nops(__start_mcount_loc, + ret = ftrace_convert_nops(NULL, + __start_mcount_loc, __stop_mcount_loc); return; @@ -1381,10 +1421,17 @@ int register_ftrace_function(struct ftrace_ops *ops) return -1; mutex_lock(&ftrace_sysctl_lock); + + if (ftrace_tracing_type == FTRACE_TYPE_RETURN) { + ret = -EBUSY; + goto out; + } + ret = __register_ftrace_function(ops); ftrace_startup(); - mutex_unlock(&ftrace_sysctl_lock); +out: + mutex_unlock(&ftrace_sysctl_lock); return ret; } @@ -1449,3 +1496,147 @@ ftrace_enable_sysctl(struct ctl_table *table, int write, return ret; } +#ifdef CONFIG_FUNCTION_RET_TRACER + +static atomic_t ftrace_retfunc_active; + +/* The callback that hooks the return of a function */ +trace_function_return_t ftrace_function_return = + (trace_function_return_t)ftrace_stub; + + +/* Try to assign a return stack array on FTRACE_RETSTACK_ALLOC_SIZE tasks. */ +static int alloc_retstack_tasklist(struct ftrace_ret_stack **ret_stack_list) +{ + int i; + int ret = 0; + unsigned long flags; + int start = 0, end = FTRACE_RETSTACK_ALLOC_SIZE; + struct task_struct *g, *t; + + for (i = 0; i < FTRACE_RETSTACK_ALLOC_SIZE; i++) { + ret_stack_list[i] = kmalloc(FTRACE_RETFUNC_DEPTH + * sizeof(struct ftrace_ret_stack), + GFP_KERNEL); + if (!ret_stack_list[i]) { + start = 0; + end = i; + ret = -ENOMEM; + goto free; + } + } + + read_lock_irqsave(&tasklist_lock, flags); + do_each_thread(g, t) { + if (start == end) { + ret = -EAGAIN; + goto unlock; + } + + if (t->ret_stack == NULL) { + t->ret_stack = ret_stack_list[start++]; + t->curr_ret_stack = -1; + atomic_set(&t->trace_overrun, 0); + } + } while_each_thread(g, t); + +unlock: + read_unlock_irqrestore(&tasklist_lock, flags); +free: + for (i = start; i < end; i++) + kfree(ret_stack_list[i]); + return ret; +} + +/* Allocate a return stack for each task */ +static int start_return_tracing(void) +{ + struct ftrace_ret_stack **ret_stack_list; + int ret; + + ret_stack_list = kmalloc(FTRACE_RETSTACK_ALLOC_SIZE * + sizeof(struct ftrace_ret_stack *), + GFP_KERNEL); + + if (!ret_stack_list) + return -ENOMEM; + + do { + ret = alloc_retstack_tasklist(ret_stack_list); + } while (ret == -EAGAIN); + + kfree(ret_stack_list); + return ret; +} + +int register_ftrace_return(trace_function_return_t func) +{ + int ret = 0; + + mutex_lock(&ftrace_sysctl_lock); + + /* + * Don't launch return tracing if normal function + * tracing is already running. + */ + if (ftrace_trace_function != ftrace_stub) { + ret = -EBUSY; + goto out; + } + atomic_inc(&ftrace_retfunc_active); + ret = start_return_tracing(); + if (ret) { + atomic_dec(&ftrace_retfunc_active); + goto out; + } + ftrace_tracing_type = FTRACE_TYPE_RETURN; + ftrace_function_return = func; + ftrace_startup(); + +out: + mutex_unlock(&ftrace_sysctl_lock); + return ret; +} + +void unregister_ftrace_return(void) +{ + mutex_lock(&ftrace_sysctl_lock); + + atomic_dec(&ftrace_retfunc_active); + ftrace_function_return = (trace_function_return_t)ftrace_stub; + ftrace_shutdown(); + /* Restore normal tracing type */ + ftrace_tracing_type = FTRACE_TYPE_ENTER; + + mutex_unlock(&ftrace_sysctl_lock); +} + +/* Allocate a return stack for newly created task */ +void ftrace_retfunc_init_task(struct task_struct *t) +{ + if (atomic_read(&ftrace_retfunc_active)) { + t->ret_stack = kmalloc(FTRACE_RETFUNC_DEPTH + * sizeof(struct ftrace_ret_stack), + GFP_KERNEL); + if (!t->ret_stack) + return; + t->curr_ret_stack = -1; + atomic_set(&t->trace_overrun, 0); + } else + t->ret_stack = NULL; +} + +void ftrace_retfunc_exit_task(struct task_struct *t) +{ + struct ftrace_ret_stack *ret_stack = t->ret_stack; + + t->ret_stack = NULL; + /* NULL must become visible to IRQs before we free it: */ + barrier(); + + kfree(ret_stack); +} +#endif + + + diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 668bbb5ef2b..e206951603c 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -18,8 +18,46 @@ #include "trace.h" -/* Global flag to disable all recording to ring buffers */ -static int ring_buffers_off __read_mostly; +/* + * A fast way to enable or disable all ring buffers is to + * call tracing_on or tracing_off. Turning off the ring buffers + * prevents all ring buffers from being recorded to. + * Turning this switch on, makes it OK to write to the + * ring buffer, if the ring buffer is enabled itself. + * + * There's three layers that must be on in order to write + * to the ring buffer. + * + * 1) This global flag must be set. + * 2) The ring buffer must be enabled for recording. + * 3) The per cpu buffer must be enabled for recording. + * + * In case of an anomaly, this global flag has a bit set that + * will permantly disable all ring buffers. + */ + +/* + * Global flag to disable all recording to ring buffers + * This has two bits: ON, DISABLED + * + * ON DISABLED + * ---- ---------- + * 0 0 : ring buffers are off + * 1 0 : ring buffers are on + * X 1 : ring buffers are permanently disabled + */ + +enum { + RB_BUFFERS_ON_BIT = 0, + RB_BUFFERS_DISABLED_BIT = 1, +}; + +enum { + RB_BUFFERS_ON = 1 << RB_BUFFERS_ON_BIT, + RB_BUFFERS_DISABLED = 1 << RB_BUFFERS_DISABLED_BIT, +}; + +static long ring_buffer_flags __read_mostly = RB_BUFFERS_ON; /** * tracing_on - enable all tracing buffers @@ -29,7 +67,7 @@ static int ring_buffers_off __read_mostly; */ void tracing_on(void) { - ring_buffers_off = 0; + set_bit(RB_BUFFERS_ON_BIT, &ring_buffer_flags); } /** @@ -42,9 +80,22 @@ void tracing_on(void) */ void tracing_off(void) { - ring_buffers_off = 1; + clear_bit(RB_BUFFERS_ON_BIT, &ring_buffer_flags); +} + +/** + * tracing_off_permanent - permanently disable ring buffers + * + * This function, once called, will disable all ring buffers + * permanenty. + */ +void tracing_off_permanent(void) +{ + set_bit(RB_BUFFERS_DISABLED_BIT, &ring_buffer_flags); } +#include "trace.h" + /* Up this if you want to test the TIME_EXTENTS and normalization */ #define DEBUG_SHIFT 0 @@ -187,7 +238,8 @@ static inline int test_time_stamp(u64 delta) struct ring_buffer_per_cpu { int cpu; struct ring_buffer *buffer; - spinlock_t lock; + spinlock_t reader_lock; /* serialize readers */ + raw_spinlock_t lock; struct lock_class_key lock_key; struct list_head pages; struct buffer_page *head_page; /* read from head */ @@ -221,32 +273,16 @@ struct ring_buffer_iter { u64 read_stamp; }; +/* buffer may be either ring_buffer or ring_buffer_per_cpu */ #define RB_WARN_ON(buffer, cond) \ - do { \ - if (unlikely(cond)) { \ - atomic_inc(&buffer->record_disabled); \ - WARN_ON(1); \ - } \ - } while (0) - -#define RB_WARN_ON_RET(buffer, cond) \ - do { \ - if (unlikely(cond)) { \ - atomic_inc(&buffer->record_disabled); \ - WARN_ON(1); \ - return -1; \ - } \ - } while (0) - -#define RB_WARN_ON_ONCE(buffer, cond) \ - do { \ - static int once; \ - if (unlikely(cond) && !once) { \ - once++; \ + ({ \ + int _____ret = unlikely(cond); \ + if (_____ret) { \ atomic_inc(&buffer->record_disabled); \ WARN_ON(1); \ } \ - } while (0) + _____ret; \ + }) /** * check_pages - integrity check of buffer pages @@ -260,14 +296,18 @@ static int rb_check_pages(struct ring_buffer_per_cpu *cpu_buffer) struct list_head *head = &cpu_buffer->pages; struct buffer_page *page, *tmp; - RB_WARN_ON_RET(cpu_buffer, head->next->prev != head); - RB_WARN_ON_RET(cpu_buffer, head->prev->next != head); + if (RB_WARN_ON(cpu_buffer, head->next->prev != head)) + return -1; + if (RB_WARN_ON(cpu_buffer, head->prev->next != head)) + return -1; list_for_each_entry_safe(page, tmp, head, list) { - RB_WARN_ON_RET(cpu_buffer, - page->list.next->prev != &page->list); - RB_WARN_ON_RET(cpu_buffer, - page->list.prev->next != &page->list); + if (RB_WARN_ON(cpu_buffer, + page->list.next->prev != &page->list)) + return -1; + if (RB_WARN_ON(cpu_buffer, + page->list.prev->next != &page->list)) + return -1; } return 0; @@ -324,7 +364,8 @@ rb_allocate_cpu_buffer(struct ring_buffer *buffer, int cpu) cpu_buffer->cpu = cpu; cpu_buffer->buffer = buffer; - spin_lock_init(&cpu_buffer->lock); + spin_lock_init(&cpu_buffer->reader_lock); + cpu_buffer->lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; INIT_LIST_HEAD(&cpu_buffer->pages); page = kzalloc_node(ALIGN(sizeof(*page), cache_line_size()), @@ -473,13 +514,15 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned nr_pages) synchronize_sched(); for (i = 0; i < nr_pages; i++) { - BUG_ON(list_empty(&cpu_buffer->pages)); + if (RB_WARN_ON(cpu_buffer, list_empty(&cpu_buffer->pages))) + return; p = cpu_buffer->pages.next; page = list_entry(p, struct buffer_page, list); list_del_init(&page->list); free_buffer_page(page); } - BUG_ON(list_empty(&cpu_buffer->pages)); + if (RB_WARN_ON(cpu_buffer, list_empty(&cpu_buffer->pages))) + return; rb_reset_cpu(cpu_buffer); @@ -501,7 +544,8 @@ rb_insert_pages(struct ring_buffer_per_cpu *cpu_buffer, synchronize_sched(); for (i = 0; i < nr_pages; i++) { - BUG_ON(list_empty(pages)); + if (RB_WARN_ON(cpu_buffer, list_empty(pages))) + return; p = pages->next; page = list_entry(p, struct buffer_page, list); list_del_init(&page->list); @@ -562,7 +606,10 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size) if (size < buffer_size) { /* easy case, just free pages */ - BUG_ON(nr_pages >= buffer->pages); + if (RB_WARN_ON(buffer, nr_pages >= buffer->pages)) { + mutex_unlock(&buffer->mutex); + return -1; + } rm_pages = buffer->pages - nr_pages; @@ -581,7 +628,11 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size) * add these pages to the cpu_buffers. Otherwise we just free * them all and return -ENOMEM; */ - BUG_ON(nr_pages <= buffer->pages); + if (RB_WARN_ON(buffer, nr_pages <= buffer->pages)) { + mutex_unlock(&buffer->mutex); + return -1; + } + new_pages = nr_pages - buffer->pages; for_each_buffer_cpu(buffer, cpu) { @@ -604,7 +655,10 @@ int ring_buffer_resize(struct ring_buffer *buffer, unsigned long size) rb_insert_pages(cpu_buffer, &pages, new_pages); } - BUG_ON(!list_empty(&pages)); + if (RB_WARN_ON(buffer, !list_empty(&pages))) { + mutex_unlock(&buffer->mutex); + return -1; + } out: buffer->pages = nr_pages; @@ -693,7 +747,8 @@ static void rb_update_overflow(struct ring_buffer_per_cpu *cpu_buffer) head += rb_event_length(event)) { event = __rb_page_index(cpu_buffer->head_page, head); - BUG_ON(rb_null_event(event)); + if (RB_WARN_ON(cpu_buffer, rb_null_event(event))) + return; /* Only count data entries */ if (event->type != RINGBUF_TYPE_DATA) continue; @@ -746,8 +801,9 @@ rb_set_commit_event(struct ring_buffer_per_cpu *cpu_buffer, addr &= PAGE_MASK; while (cpu_buffer->commit_page->page != (void *)addr) { - RB_WARN_ON(cpu_buffer, - cpu_buffer->commit_page == cpu_buffer->tail_page); + if (RB_WARN_ON(cpu_buffer, + cpu_buffer->commit_page == cpu_buffer->tail_page)) + return; cpu_buffer->commit_page->commit = cpu_buffer->commit_page->write; rb_inc_page(cpu_buffer, &cpu_buffer->commit_page); @@ -894,7 +950,8 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, if (write > BUF_PAGE_SIZE) { struct buffer_page *next_page = tail_page; - spin_lock_irqsave(&cpu_buffer->lock, flags); + local_irq_save(flags); + __raw_spin_lock(&cpu_buffer->lock); rb_inc_page(cpu_buffer, &next_page); @@ -902,7 +959,8 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, reader_page = cpu_buffer->reader_page; /* we grabbed the lock before incrementing */ - RB_WARN_ON(cpu_buffer, next_page == reader_page); + if (RB_WARN_ON(cpu_buffer, next_page == reader_page)) + goto out_unlock; /* * If for some reason, we had an interrupt storm that made @@ -970,7 +1028,8 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, rb_set_commit_to_write(cpu_buffer); } - spin_unlock_irqrestore(&cpu_buffer->lock, flags); + __raw_spin_unlock(&cpu_buffer->lock); + local_irq_restore(flags); /* fail and let the caller try again */ return ERR_PTR(-EAGAIN); @@ -978,7 +1037,8 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, /* We reserved something on the buffer */ - BUG_ON(write > BUF_PAGE_SIZE); + if (RB_WARN_ON(cpu_buffer, write > BUF_PAGE_SIZE)) + return NULL; event = __rb_page_index(tail_page, tail); rb_update_event(event, type, length); @@ -993,7 +1053,8 @@ __rb_reserve_next(struct ring_buffer_per_cpu *cpu_buffer, return event; out_unlock: - spin_unlock_irqrestore(&cpu_buffer->lock, flags); + __raw_spin_unlock(&cpu_buffer->lock); + local_irq_restore(flags); return NULL; } @@ -1076,10 +1137,8 @@ rb_reserve_next_event(struct ring_buffer_per_cpu *cpu_buffer, * storm or we have something buggy. * Bail! */ - if (unlikely(++nr_loops > 1000)) { - RB_WARN_ON(cpu_buffer, 1); + if (RB_WARN_ON(cpu_buffer, ++nr_loops > 1000)) return NULL; - } ts = ring_buffer_time_stamp(cpu_buffer->cpu); @@ -1175,15 +1234,14 @@ ring_buffer_lock_reserve(struct ring_buffer *buffer, struct ring_buffer_event *event; int cpu, resched; - if (ring_buffers_off) + if (ring_buffer_flags != RB_BUFFERS_ON) return NULL; if (atomic_read(&buffer->record_disabled)) return NULL; /* If we are tracing schedule, we don't want to recurse */ - resched = need_resched(); - preempt_disable_notrace(); + resched = ftrace_preempt_disable(); cpu = raw_smp_processor_id(); @@ -1214,10 +1272,7 @@ ring_buffer_lock_reserve(struct ring_buffer *buffer, return event; out: - if (resched) - preempt_enable_no_resched_notrace(); - else - preempt_enable_notrace(); + ftrace_preempt_enable(resched); return NULL; } @@ -1259,12 +1314,9 @@ int ring_buffer_unlock_commit(struct ring_buffer *buffer, /* * Only the last preempt count needs to restore preemption. */ - if (preempt_count() == 1) { - if (per_cpu(rb_need_resched, cpu)) - preempt_enable_no_resched_notrace(); - else - preempt_enable_notrace(); - } else + if (preempt_count() == 1) + ftrace_preempt_enable(per_cpu(rb_need_resched, cpu)); + else preempt_enable_no_resched_notrace(); return 0; @@ -1294,14 +1346,13 @@ int ring_buffer_write(struct ring_buffer *buffer, int ret = -EBUSY; int cpu, resched; - if (ring_buffers_off) + if (ring_buffer_flags != RB_BUFFERS_ON) return -EBUSY; if (atomic_read(&buffer->record_disabled)) return -EBUSY; - resched = need_resched(); - preempt_disable_notrace(); + resched = ftrace_preempt_disable(); cpu = raw_smp_processor_id(); @@ -1327,10 +1378,7 @@ int ring_buffer_write(struct ring_buffer *buffer, ret = 0; out: - if (resched) - preempt_enable_no_resched_notrace(); - else - preempt_enable_notrace(); + ftrace_preempt_enable(resched); return ret; } @@ -1489,14 +1537,7 @@ unsigned long ring_buffer_overruns(struct ring_buffer *buffer) return overruns; } -/** - * ring_buffer_iter_reset - reset an iterator - * @iter: The iterator to reset - * - * Resets the iterator, so that it will start from the beginning - * again. - */ -void ring_buffer_iter_reset(struct ring_buffer_iter *iter) +static void rb_iter_reset(struct ring_buffer_iter *iter) { struct ring_buffer_per_cpu *cpu_buffer = iter->cpu_buffer; @@ -1515,6 +1556,23 @@ void ring_buffer_iter_reset(struct ring_buffer_iter *iter) } /** + * ring_buffer_iter_reset - reset an iterator + * @iter: The iterator to reset + * + * Resets the iterator, so that it will start from the beginning + * again. + */ +void ring_buffer_iter_reset(struct ring_buffer_iter *iter) +{ + struct ring_buffer_per_cpu *cpu_buffer = iter->cpu_buffer; + unsigned long flags; + + spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + rb_iter_reset(iter); + spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); +} + +/** * ring_buffer_iter_empty - check if an iterator has no more to read * @iter: The iterator to check */ @@ -1597,7 +1655,8 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) unsigned long flags; int nr_loops = 0; - spin_lock_irqsave(&cpu_buffer->lock, flags); + local_irq_save(flags); + __raw_spin_lock(&cpu_buffer->lock); again: /* @@ -1606,8 +1665,7 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) * a case where we will loop three times. There should be no * reason to loop four times (that I know of). */ - if (unlikely(++nr_loops > 3)) { - RB_WARN_ON(cpu_buffer, 1); + if (RB_WARN_ON(cpu_buffer, ++nr_loops > 3)) { reader = NULL; goto out; } @@ -1619,8 +1677,9 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) goto out; /* Never should we have an index greater than the size */ - RB_WARN_ON(cpu_buffer, - cpu_buffer->reader_page->read > rb_page_size(reader)); + if (RB_WARN_ON(cpu_buffer, + cpu_buffer->reader_page->read > rb_page_size(reader))) + goto out; /* check if we caught up to the tail */ reader = NULL; @@ -1659,7 +1718,8 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) goto again; out: - spin_unlock_irqrestore(&cpu_buffer->lock, flags); + __raw_spin_unlock(&cpu_buffer->lock); + local_irq_restore(flags); return reader; } @@ -1673,7 +1733,8 @@ static void rb_advance_reader(struct ring_buffer_per_cpu *cpu_buffer) reader = rb_get_reader_page(cpu_buffer); /* This function should not be called when buffer is empty */ - BUG_ON(!reader); + if (RB_WARN_ON(cpu_buffer, !reader)) + return; event = rb_reader_event(cpu_buffer); @@ -1700,7 +1761,9 @@ static void rb_advance_iter(struct ring_buffer_iter *iter) * Check if we are at the end of the buffer. */ if (iter->head >= rb_page_size(iter->head_page)) { - BUG_ON(iter->head_page == cpu_buffer->commit_page); + if (RB_WARN_ON(buffer, + iter->head_page == cpu_buffer->commit_page)) + return; rb_inc_iter(iter); return; } @@ -1713,8 +1776,10 @@ static void rb_advance_iter(struct ring_buffer_iter *iter) * This should not be called to advance the header if we are * at the tail of the buffer. */ - BUG_ON((iter->head_page == cpu_buffer->commit_page) && - (iter->head + length > rb_commit_index(cpu_buffer))); + if (RB_WARN_ON(cpu_buffer, + (iter->head_page == cpu_buffer->commit_page) && + (iter->head + length > rb_commit_index(cpu_buffer)))) + return; rb_update_iter_read_stamp(iter, event); @@ -1726,17 +1791,8 @@ static void rb_advance_iter(struct ring_buffer_iter *iter) rb_advance_iter(iter); } -/** - * ring_buffer_peek - peek at the next event to be read - * @buffer: The ring buffer to read - * @cpu: The cpu to peak at - * @ts: The timestamp counter of this event. - * - * This will return the event that will be read next, but does - * not consume the data. - */ -struct ring_buffer_event * -ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts) +static struct ring_buffer_event * +rb_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts) { struct ring_buffer_per_cpu *cpu_buffer; struct ring_buffer_event *event; @@ -1757,10 +1813,8 @@ ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts) * can have. Nesting 10 deep of interrupts is clearly * an anomaly. */ - if (unlikely(++nr_loops > 10)) { - RB_WARN_ON(cpu_buffer, 1); + if (RB_WARN_ON(cpu_buffer, ++nr_loops > 10)) return NULL; - } reader = rb_get_reader_page(cpu_buffer); if (!reader) @@ -1798,16 +1852,8 @@ ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts) return NULL; } -/** - * ring_buffer_iter_peek - peek at the next event to be read - * @iter: The ring buffer iterator - * @ts: The timestamp counter of this event. - * - * This will return the event that will be read next, but does - * not increment the iterator. - */ -struct ring_buffer_event * -ring_buffer_iter_peek(struct ring_buffer_iter *iter, u64 *ts) +static struct ring_buffer_event * +rb_iter_peek(struct ring_buffer_iter *iter, u64 *ts) { struct ring_buffer *buffer; struct ring_buffer_per_cpu *cpu_buffer; @@ -1829,10 +1875,8 @@ ring_buffer_iter_peek(struct ring_buffer_iter *iter, u64 *ts) * can have. Nesting 10 deep of interrupts is clearly * an anomaly. */ - if (unlikely(++nr_loops > 10)) { - RB_WARN_ON(cpu_buffer, 1); + if (RB_WARN_ON(cpu_buffer, ++nr_loops > 10)) return NULL; - } if (rb_per_cpu_empty(cpu_buffer)) return NULL; @@ -1869,6 +1913,51 @@ ring_buffer_iter_peek(struct ring_buffer_iter *iter, u64 *ts) } /** + * ring_buffer_peek - peek at the next event to be read + * @buffer: The ring buffer to read + * @cpu: The cpu to peak at + * @ts: The timestamp counter of this event. + * + * This will return the event that will be read next, but does + * not consume the data. + */ +struct ring_buffer_event * +ring_buffer_peek(struct ring_buffer *buffer, int cpu, u64 *ts) +{ + struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; + struct ring_buffer_event *event; + unsigned long flags; + + spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + event = rb_buffer_peek(buffer, cpu, ts); + spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + + return event; +} + +/** + * ring_buffer_iter_peek - peek at the next event to be read + * @iter: The ring buffer iterator + * @ts: The timestamp counter of this event. + * + * This will return the event that will be read next, but does + * not increment the iterator. + */ +struct ring_buffer_event * +ring_buffer_iter_peek(struct ring_buffer_iter *iter, u64 *ts) +{ + struct ring_buffer_per_cpu *cpu_buffer = iter->cpu_buffer; + struct ring_buffer_event *event; + unsigned long flags; + + spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + event = rb_iter_peek(iter, ts); + spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + + return event; +} + +/** * ring_buffer_consume - return an event and consume it * @buffer: The ring buffer to get the next event from * @@ -1879,19 +1968,24 @@ ring_buffer_iter_peek(struct ring_buffer_iter *iter, u64 *ts) struct ring_buffer_event * ring_buffer_consume(struct ring_buffer *buffer, int cpu, u64 *ts) { - struct ring_buffer_per_cpu *cpu_buffer; + struct ring_buffer_per_cpu *cpu_buffer = buffer->buffers[cpu]; struct ring_buffer_event *event; + unsigned long flags; if (!cpu_isset(cpu, buffer->cpumask)) return NULL; - event = ring_buffer_peek(buffer, cpu, ts); + spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + + event = rb_buffer_peek(buffer, cpu, ts); if (!event) - return NULL; + goto out; - cpu_buffer = buffer->buffers[cpu]; rb_advance_reader(cpu_buffer); + out: + spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); + return event; } @@ -1928,9 +2022,11 @@ ring_buffer_read_start(struct ring_buffer *buffer, int cpu) atomic_inc(&cpu_buffer->record_disabled); synchronize_sched(); - spin_lock_irqsave(&cpu_buffer->lock, flags); - ring_buffer_iter_reset(iter); - spin_unlock_irqrestore(&cpu_buffer->lock, flags); + spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + __raw_spin_lock(&cpu_buffer->lock); + rb_iter_reset(iter); + __raw_spin_unlock(&cpu_buffer->lock); + spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); return iter; } @@ -1962,12 +2058,17 @@ struct ring_buffer_event * ring_buffer_read(struct ring_buffer_iter *iter, u64 *ts) { struct ring_buffer_event *event; + struct ring_buffer_per_cpu *cpu_buffer = iter->cpu_buffer; + unsigned long flags; - event = ring_buffer_iter_peek(iter, ts); + spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + event = rb_iter_peek(iter, ts); if (!event) - return NULL; + goto out; rb_advance_iter(iter); + out: + spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); return event; } @@ -2016,11 +2117,15 @@ void ring_buffer_reset_cpu(struct ring_buffer *buffer, int cpu) if (!cpu_isset(cpu, buffer->cpumask)) return; - spin_lock_irqsave(&cpu_buffer->lock, flags); + spin_lock_irqsave(&cpu_buffer->reader_lock, flags); + + __raw_spin_lock(&cpu_buffer->lock); rb_reset_cpu(cpu_buffer); - spin_unlock_irqrestore(&cpu_buffer->lock, flags); + __raw_spin_unlock(&cpu_buffer->lock); + + spin_unlock_irqrestore(&cpu_buffer->reader_lock, flags); } /** @@ -2122,12 +2227,14 @@ static ssize_t rb_simple_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - int *p = filp->private_data; + long *p = filp->private_data; char buf[64]; int r; - /* !ring_buffers_off == tracing_on */ - r = sprintf(buf, "%d\n", !*p); + if (test_bit(RB_BUFFERS_DISABLED_BIT, p)) + r = sprintf(buf, "permanently disabled\n"); + else + r = sprintf(buf, "%d\n", test_bit(RB_BUFFERS_ON_BIT, p)); return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); } @@ -2136,7 +2243,7 @@ static ssize_t rb_simple_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - int *p = filp->private_data; + long *p = filp->private_data; char buf[64]; long val; int ret; @@ -2153,8 +2260,10 @@ rb_simple_write(struct file *filp, const char __user *ubuf, if (ret < 0) return ret; - /* !ring_buffers_off == tracing_on */ - *p = !val; + if (val) + set_bit(RB_BUFFERS_ON_BIT, p); + else + clear_bit(RB_BUFFERS_ON_BIT, p); (*ppos)++; @@ -2176,7 +2285,7 @@ static __init int rb_init_debugfs(void) d_tracer = tracing_init_dentry(); entry = debugfs_create_file("tracing_on", 0644, d_tracer, - &ring_buffers_off, &rb_simple_fops); + &ring_buffer_flags, &rb_simple_fops); if (!entry) pr_warning("Could not create debugfs 'tracing_on' entry\n"); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index d86e3252f30..a45b59e53fb 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -30,6 +30,7 @@ #include <linux/gfp.h> #include <linux/fs.h> #include <linux/kprobes.h> +#include <linux/seq_file.h> #include <linux/writeback.h> #include <linux/stacktrace.h> @@ -43,6 +44,29 @@ unsigned long __read_mostly tracing_max_latency = (cycle_t)ULONG_MAX; unsigned long __read_mostly tracing_thresh; +/* For tracers that don't implement custom flags */ +static struct tracer_opt dummy_tracer_opt[] = { + { } +}; + +static struct tracer_flags dummy_tracer_flags = { + .val = 0, + .opts = dummy_tracer_opt +}; + +static int dummy_set_flag(u32 old_flags, u32 bit, int set) +{ + return 0; +} + +/* + * Kill all tracing for good (never come back). + * It is initialized to 1 but will turn to zero if the initialization + * of the tracer is successful. But that is the only place that sets + * this back to zero. + */ +int tracing_disabled = 1; + static DEFINE_PER_CPU(local_t, ftrace_cpu_disabled); static inline void ftrace_disable_cpu(void) @@ -62,7 +86,36 @@ static cpumask_t __read_mostly tracing_buffer_mask; #define for_each_tracing_cpu(cpu) \ for_each_cpu_mask(cpu, tracing_buffer_mask) -static int tracing_disabled = 1; +/* + * ftrace_dump_on_oops - variable to dump ftrace buffer on oops + * + * If there is an oops (or kernel panic) and the ftrace_dump_on_oops + * is set, then ftrace_dump is called. This will output the contents + * of the ftrace buffers to the console. This is very useful for + * capturing traces that lead to crashes and outputing it to a + * serial console. + * + * It is default off, but you can enable it with either specifying + * "ftrace_dump_on_oops" in the kernel command line, or setting + * /proc/sys/kernel/ftrace_dump_on_oops to true. + */ +int ftrace_dump_on_oops; + +static int tracing_set_tracer(char *buf); + +static int __init set_ftrace(char *str) +{ + tracing_set_tracer(str); + return 1; +} +__setup("ftrace", set_ftrace); + +static int __init set_ftrace_dump_on_oops(char *str) +{ + ftrace_dump_on_oops = 1; + return 1; +} +__setup("ftrace_dump_on_oops", set_ftrace_dump_on_oops); long ns2usecs(cycle_t nsec) @@ -112,6 +165,19 @@ static DEFINE_PER_CPU(struct trace_array_cpu, max_data); /* tracer_enabled is used to toggle activation of a tracer */ static int tracer_enabled = 1; +/** + * tracing_is_enabled - return tracer_enabled status + * + * This function is used by other tracers to know the status + * of the tracer_enabled flag. Tracers may use this function + * to know if it should enable their features when starting + * up. See irqsoff tracer for an example (start_irqsoff_tracer). + */ +int tracing_is_enabled(void) +{ + return tracer_enabled; +} + /* function tracing enabled */ int ftrace_function_enabled; @@ -153,8 +219,9 @@ static DEFINE_MUTEX(trace_types_lock); /* trace_wait is a waitqueue for tasks blocked on trace_poll */ static DECLARE_WAIT_QUEUE_HEAD(trace_wait); -/* trace_flags holds iter_ctrl options */ -unsigned long trace_flags = TRACE_ITER_PRINT_PARENT; +/* trace_flags holds trace_options default values */ +unsigned long trace_flags = TRACE_ITER_PRINT_PARENT | TRACE_ITER_PRINTK | + TRACE_ITER_ANNOTATE; /** * trace_wake_up - wake up tasks waiting for trace input @@ -193,13 +260,6 @@ unsigned long nsecs_to_usecs(unsigned long nsecs) return nsecs / 1000; } -/* - * TRACE_ITER_SYM_MASK masks the options in trace_flags that - * control the output of kernel symbols. - */ -#define TRACE_ITER_SYM_MASK \ - (TRACE_ITER_PRINT_PARENT|TRACE_ITER_SYM_OFFSET|TRACE_ITER_SYM_ADDR) - /* These must match the bit postions in trace_iterator_flags */ static const char *trace_options[] = { "print-parent", @@ -213,6 +273,11 @@ static const char *trace_options[] = { "stacktrace", "sched-tree", "ftrace_printk", + "ftrace_preempt", + "branch", + "annotate", + "userstacktrace", + "sym-userobj", NULL }; @@ -359,6 +424,28 @@ trace_seq_putmem_hex(struct trace_seq *s, void *mem, size_t len) return trace_seq_putmem(s, hex, j); } +static int +trace_seq_path(struct trace_seq *s, struct path *path) +{ + unsigned char *p; + + if (s->len >= (PAGE_SIZE - 1)) + return 0; + p = d_path(path, s->buffer + s->len, PAGE_SIZE - s->len); + if (!IS_ERR(p)) { + p = mangle_path(s->buffer + s->len, p, "\n"); + if (p) { + s->len = p - s->buffer; + return 1; + } + } else { + s->buffer[s->len++] = '?'; + return 1; + } + + return 0; +} + static void trace_seq_reset(struct trace_seq *s) { @@ -470,7 +557,15 @@ int register_tracer(struct tracer *type) return -1; } + /* + * When this gets called we hold the BKL which means that + * preemption is disabled. Various trace selftests however + * need to disable and enable preemption for successful tests. + * So we drop the BKL here and grab it after the tests again. + */ + unlock_kernel(); mutex_lock(&trace_types_lock); + for (t = trace_types; t; t = t->next) { if (strcmp(type->name, t->name) == 0) { /* already found */ @@ -481,11 +576,18 @@ int register_tracer(struct tracer *type) } } + if (!type->set_flag) + type->set_flag = &dummy_set_flag; + if (!type->flags) + type->flags = &dummy_tracer_flags; + else + if (!type->flags->opts) + type->flags->opts = dummy_tracer_opt; + #ifdef CONFIG_FTRACE_STARTUP_TEST if (type->selftest) { struct tracer *saved_tracer = current_trace; struct trace_array *tr = &global_trace; - int saved_ctrl = tr->ctrl; int i; /* * Run a selftest on this tracer. @@ -494,25 +596,23 @@ int register_tracer(struct tracer *type) * internal tracing to verify that everything is in order. * If we fail, we do not register this tracer. */ - for_each_tracing_cpu(i) { + for_each_tracing_cpu(i) tracing_reset(tr, i); - } + current_trace = type; - tr->ctrl = 0; /* the test is responsible for initializing and enabling */ pr_info("Testing tracer %s: ", type->name); ret = type->selftest(type, tr); /* the test is responsible for resetting too */ current_trace = saved_tracer; - tr->ctrl = saved_ctrl; if (ret) { printk(KERN_CONT "FAILED!\n"); goto out; } /* Only reset on passing, to avoid touching corrupted buffers */ - for_each_tracing_cpu(i) { + for_each_tracing_cpu(i) tracing_reset(tr, i); - } + printk(KERN_CONT "PASSED\n"); } #endif @@ -525,6 +625,7 @@ int register_tracer(struct tracer *type) out: mutex_unlock(&trace_types_lock); + lock_kernel(); return ret; } @@ -581,6 +682,91 @@ static void trace_init_cmdlines(void) cmdline_idx = 0; } +static int trace_stop_count; +static DEFINE_SPINLOCK(tracing_start_lock); + +/** + * ftrace_off_permanent - disable all ftrace code permanently + * + * This should only be called when a serious anomally has + * been detected. This will turn off the function tracing, + * ring buffers, and other tracing utilites. It takes no + * locks and can be called from any context. + */ +void ftrace_off_permanent(void) +{ + tracing_disabled = 1; + ftrace_stop(); + tracing_off_permanent(); +} + +/** + * tracing_start - quick start of the tracer + * + * If tracing is enabled but was stopped by tracing_stop, + * this will start the tracer back up. + */ +void tracing_start(void) +{ + struct ring_buffer *buffer; + unsigned long flags; + + if (tracing_disabled) + return; + + spin_lock_irqsave(&tracing_start_lock, flags); + if (--trace_stop_count) + goto out; + + if (trace_stop_count < 0) { + /* Someone screwed up their debugging */ + WARN_ON_ONCE(1); + trace_stop_count = 0; + goto out; + } + + + buffer = global_trace.buffer; + if (buffer) + ring_buffer_record_enable(buffer); + + buffer = max_tr.buffer; + if (buffer) + ring_buffer_record_enable(buffer); + + ftrace_start(); + out: + spin_unlock_irqrestore(&tracing_start_lock, flags); +} + +/** + * tracing_stop - quick stop of the tracer + * + * Light weight way to stop tracing. Use in conjunction with + * tracing_start. + */ +void tracing_stop(void) +{ + struct ring_buffer *buffer; + unsigned long flags; + + ftrace_stop(); + spin_lock_irqsave(&tracing_start_lock, flags); + if (trace_stop_count++) + goto out; + + buffer = global_trace.buffer; + if (buffer) + ring_buffer_record_disable(buffer); + + buffer = max_tr.buffer; + if (buffer) + ring_buffer_record_disable(buffer); + + out: + spin_unlock_irqrestore(&tracing_start_lock, flags); +} + void trace_stop_cmdline_recording(void); static void trace_save_cmdline(struct task_struct *tsk) @@ -655,6 +841,7 @@ tracing_generic_entry_update(struct trace_entry *entry, unsigned long flags, entry->preempt_count = pc & 0xff; entry->pid = (tsk) ? tsk->pid : 0; + entry->tgid = (tsk) ? tsk->tgid : 0; entry->flags = #ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT (irqs_disabled_flags(flags) ? TRACE_FLAG_IRQS_OFF : 0) | @@ -691,6 +878,36 @@ trace_function(struct trace_array *tr, struct trace_array_cpu *data, ring_buffer_unlock_commit(tr->buffer, event, irq_flags); } +#ifdef CONFIG_FUNCTION_RET_TRACER +static void __trace_function_return(struct trace_array *tr, + struct trace_array_cpu *data, + struct ftrace_retfunc *trace, + unsigned long flags, + int pc) +{ + struct ring_buffer_event *event; + struct ftrace_ret_entry *entry; + unsigned long irq_flags; + + if (unlikely(local_read(&__get_cpu_var(ftrace_cpu_disabled)))) + return; + + event = ring_buffer_lock_reserve(global_trace.buffer, sizeof(*entry), + &irq_flags); + if (!event) + return; + entry = ring_buffer_event_data(event); + tracing_generic_entry_update(&entry->ent, flags, pc); + entry->ent.type = TRACE_FN_RET; + entry->ip = trace->func; + entry->parent_ip = trace->ret; + entry->rettime = trace->rettime; + entry->calltime = trace->calltime; + entry->overrun = trace->overrun; + ring_buffer_unlock_commit(global_trace.buffer, event, irq_flags); +} +#endif + void ftrace(struct trace_array *tr, struct trace_array_cpu *data, unsigned long ip, unsigned long parent_ip, unsigned long flags, @@ -742,6 +959,44 @@ void __trace_stack(struct trace_array *tr, ftrace_trace_stack(tr, data, flags, skip, preempt_count()); } +static void ftrace_trace_userstack(struct trace_array *tr, + struct trace_array_cpu *data, + unsigned long flags, int pc) +{ + struct ring_buffer_event *event; + struct userstack_entry *entry; + struct stack_trace trace; + unsigned long irq_flags; + + if (!(trace_flags & TRACE_ITER_USERSTACKTRACE)) + return; + + event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), + &irq_flags); + if (!event) + return; + entry = ring_buffer_event_data(event); + tracing_generic_entry_update(&entry->ent, flags, pc); + entry->ent.type = TRACE_USER_STACK; + + memset(&entry->caller, 0, sizeof(entry->caller)); + + trace.nr_entries = 0; + trace.max_entries = FTRACE_STACK_ENTRIES; + trace.skip = 0; + trace.entries = entry->caller; + + save_stack_trace_user(&trace); + ring_buffer_unlock_commit(tr->buffer, event, irq_flags); +} + +void __trace_userstack(struct trace_array *tr, + struct trace_array_cpu *data, + unsigned long flags) +{ + ftrace_trace_userstack(tr, data, flags, preempt_count()); +} + static void ftrace_trace_special(void *__tr, void *__data, unsigned long arg1, unsigned long arg2, unsigned long arg3, @@ -765,6 +1020,7 @@ ftrace_trace_special(void *__tr, void *__data, entry->arg3 = arg3; ring_buffer_unlock_commit(tr->buffer, event, irq_flags); ftrace_trace_stack(tr, data, irq_flags, 4, pc); + ftrace_trace_userstack(tr, data, irq_flags, pc); trace_wake_up(); } @@ -803,6 +1059,7 @@ tracing_sched_switch_trace(struct trace_array *tr, entry->next_cpu = task_cpu(next); ring_buffer_unlock_commit(tr->buffer, event, irq_flags); ftrace_trace_stack(tr, data, flags, 5, pc); + ftrace_trace_userstack(tr, data, flags, pc); } void @@ -832,6 +1089,7 @@ tracing_sched_wakeup_trace(struct trace_array *tr, entry->next_cpu = task_cpu(wakee); ring_buffer_unlock_commit(tr->buffer, event, irq_flags); ftrace_trace_stack(tr, data, flags, 6, pc); + ftrace_trace_userstack(tr, data, flags, pc); trace_wake_up(); } @@ -841,26 +1099,28 @@ ftrace_special(unsigned long arg1, unsigned long arg2, unsigned long arg3) { struct trace_array *tr = &global_trace; struct trace_array_cpu *data; + unsigned long flags; int cpu; int pc; - if (tracing_disabled || !tr->ctrl) + if (tracing_disabled) return; pc = preempt_count(); - preempt_disable_notrace(); + local_irq_save(flags); cpu = raw_smp_processor_id(); data = tr->data[cpu]; - if (likely(!atomic_read(&data->disabled))) + if (likely(atomic_inc_return(&data->disabled) == 1)) ftrace_trace_special(tr, data, arg1, arg2, arg3, pc); - preempt_enable_notrace(); + atomic_dec(&data->disabled); + local_irq_restore(flags); } #ifdef CONFIG_FUNCTION_TRACER static void -function_trace_call(unsigned long ip, unsigned long parent_ip) +function_trace_call_preempt_only(unsigned long ip, unsigned long parent_ip) { struct trace_array *tr = &global_trace; struct trace_array_cpu *data; @@ -873,8 +1133,7 @@ function_trace_call(unsigned long ip, unsigned long parent_ip) return; pc = preempt_count(); - resched = need_resched(); - preempt_disable_notrace(); + resched = ftrace_preempt_disable(); local_save_flags(flags); cpu = raw_smp_processor_id(); data = tr->data[cpu]; @@ -884,11 +1143,62 @@ function_trace_call(unsigned long ip, unsigned long parent_ip) trace_function(tr, data, ip, parent_ip, flags, pc); atomic_dec(&data->disabled); - if (resched) - preempt_enable_no_resched_notrace(); - else - preempt_enable_notrace(); + ftrace_preempt_enable(resched); +} + +static void +function_trace_call(unsigned long ip, unsigned long parent_ip) +{ + struct trace_array *tr = &global_trace; + struct trace_array_cpu *data; + unsigned long flags; + long disabled; + int cpu; + int pc; + + if (unlikely(!ftrace_function_enabled)) + return; + + /* + * Need to use raw, since this must be called before the + * recursive protection is performed. + */ + local_irq_save(flags); + cpu = raw_smp_processor_id(); + data = tr->data[cpu]; + disabled = atomic_inc_return(&data->disabled); + + if (likely(disabled == 1)) { + pc = preempt_count(); + trace_function(tr, data, ip, parent_ip, flags, pc); + } + + atomic_dec(&data->disabled); + local_irq_restore(flags); +} + +#ifdef CONFIG_FUNCTION_RET_TRACER +void trace_function_return(struct ftrace_retfunc *trace) +{ + struct trace_array *tr = &global_trace; + struct trace_array_cpu *data; + unsigned long flags; + long disabled; + int cpu; + int pc; + + raw_local_irq_save(flags); + cpu = raw_smp_processor_id(); + data = tr->data[cpu]; + disabled = atomic_inc_return(&data->disabled); + if (likely(disabled == 1)) { + pc = preempt_count(); + __trace_function_return(tr, data, trace, flags, pc); + } + atomic_dec(&data->disabled); + raw_local_irq_restore(flags); } +#endif /* CONFIG_FUNCTION_RET_TRACER */ static struct ftrace_ops trace_ops __read_mostly = { @@ -898,9 +1208,14 @@ static struct ftrace_ops trace_ops __read_mostly = void tracing_start_function_trace(void) { ftrace_function_enabled = 0; + + if (trace_flags & TRACE_ITER_PREEMPTONLY) + trace_ops.func = function_trace_call_preempt_only; + else + trace_ops.func = function_trace_call; + register_ftrace_function(&trace_ops); - if (tracer_enabled) - ftrace_function_enabled = 1; + ftrace_function_enabled = 1; } void tracing_stop_function_trace(void) @@ -912,6 +1227,7 @@ void tracing_stop_function_trace(void) enum trace_file_type { TRACE_FILE_LAT_FMT = 1, + TRACE_FILE_ANNOTATE = 2, }; static void trace_iterator_increment(struct trace_iterator *iter, int cpu) @@ -1047,10 +1363,6 @@ static void *s_start(struct seq_file *m, loff_t *pos) atomic_inc(&trace_record_cmdline_disabled); - /* let the tracer grab locks here if needed */ - if (current_trace->start) - current_trace->start(iter); - if (*pos != iter->pos) { iter->ent = NULL; iter->cpu = 0; @@ -1077,14 +1389,7 @@ static void *s_start(struct seq_file *m, loff_t *pos) static void s_stop(struct seq_file *m, void *p) { - struct trace_iterator *iter = m->private; - atomic_dec(&trace_record_cmdline_disabled); - - /* let the tracer release locks here if needed */ - if (current_trace && current_trace == iter->trace && iter->trace->stop) - iter->trace->stop(iter); - mutex_unlock(&trace_types_lock); } @@ -1143,7 +1448,7 @@ seq_print_sym_offset(struct trace_seq *s, const char *fmt, # define IP_FMT "%016lx" #endif -static int +int seq_print_ip_sym(struct trace_seq *s, unsigned long ip, unsigned long sym_flags) { int ret; @@ -1164,6 +1469,78 @@ seq_print_ip_sym(struct trace_seq *s, unsigned long ip, unsigned long sym_flags) return ret; } +static inline int seq_print_user_ip(struct trace_seq *s, struct mm_struct *mm, + unsigned long ip, unsigned long sym_flags) +{ + struct file *file = NULL; + unsigned long vmstart = 0; + int ret = 1; + + if (mm) { + const struct vm_area_struct *vma; + + down_read(&mm->mmap_sem); + vma = find_vma(mm, ip); + if (vma) { + file = vma->vm_file; + vmstart = vma->vm_start; + } + if (file) { + ret = trace_seq_path(s, &file->f_path); + if (ret) + ret = trace_seq_printf(s, "[+0x%lx]", ip - vmstart); + } + up_read(&mm->mmap_sem); + } + if (ret && ((sym_flags & TRACE_ITER_SYM_ADDR) || !file)) + ret = trace_seq_printf(s, " <" IP_FMT ">", ip); + return ret; +} + +static int +seq_print_userip_objs(const struct userstack_entry *entry, struct trace_seq *s, + unsigned long sym_flags) +{ + struct mm_struct *mm = NULL; + int ret = 1; + unsigned int i; + + if (trace_flags & TRACE_ITER_SYM_USEROBJ) { + struct task_struct *task; + /* + * we do the lookup on the thread group leader, + * since individual threads might have already quit! + */ + rcu_read_lock(); + task = find_task_by_vpid(entry->ent.tgid); + if (task) + mm = get_task_mm(task); + rcu_read_unlock(); + } + + for (i = 0; i < FTRACE_STACK_ENTRIES; i++) { + unsigned long ip = entry->caller[i]; + + if (ip == ULONG_MAX || !ret) + break; + if (i && ret) + ret = trace_seq_puts(s, " <- "); + if (!ip) { + if (ret) + ret = trace_seq_puts(s, "??"); + continue; + } + if (!ret) + break; + if (ret) + ret = seq_print_user_ip(s, mm, ip, sym_flags); + } + + if (mm) + mmput(mm); + return ret; +} + static void print_lat_help_header(struct seq_file *m) { seq_puts(m, "# _------=> CPU# \n"); @@ -1338,6 +1715,23 @@ void trace_seq_print_cont(struct trace_seq *s, struct trace_iterator *iter) trace_seq_putc(s, '\n'); } +static void test_cpu_buff_start(struct trace_iterator *iter) +{ + struct trace_seq *s = &iter->seq; + + if (!(trace_flags & TRACE_ITER_ANNOTATE)) + return; + + if (!(iter->iter_flags & TRACE_FILE_ANNOTATE)) + return; + + if (cpu_isset(iter->cpu, iter->started)) + return; + + cpu_set(iter->cpu, iter->started); + trace_seq_printf(s, "##### CPU %u buffer started ####\n", iter->cpu); +} + static enum print_line_t print_lat_fmt(struct trace_iterator *iter, unsigned int trace_idx, int cpu) { @@ -1357,6 +1751,8 @@ print_lat_fmt(struct trace_iterator *iter, unsigned int trace_idx, int cpu) if (entry->type == TRACE_CONT) return TRACE_TYPE_HANDLED; + test_cpu_buff_start(iter); + next_entry = find_next_entry(iter, NULL, &next_ts); if (!next_entry) next_ts = iter->ts; @@ -1448,6 +1844,27 @@ print_lat_fmt(struct trace_iterator *iter, unsigned int trace_idx, int cpu) trace_seq_print_cont(s, iter); break; } + case TRACE_BRANCH: { + struct trace_branch *field; + + trace_assign_type(field, entry); + + trace_seq_printf(s, "[%s] %s:%s:%d\n", + field->correct ? " ok " : " MISS ", + field->func, + field->file, + field->line); + break; + } + case TRACE_USER_STACK: { + struct userstack_entry *field; + + trace_assign_type(field, entry); + + seq_print_userip_objs(field, s, sym_flags); + trace_seq_putc(s, '\n'); + break; + } default: trace_seq_printf(s, "Unknown type %d\n", entry->type); } @@ -1472,6 +1889,8 @@ static enum print_line_t print_trace_fmt(struct trace_iterator *iter) if (entry->type == TRACE_CONT) return TRACE_TYPE_HANDLED; + test_cpu_buff_start(iter); + comm = trace_find_cmdline(iter->ent->pid); t = ns2usecs(iter->ts); @@ -1581,6 +2000,35 @@ static enum print_line_t print_trace_fmt(struct trace_iterator *iter) trace_seq_print_cont(s, iter); break; } + case TRACE_FN_RET: { + return print_return_function(iter); + break; + } + case TRACE_BRANCH: { + struct trace_branch *field; + + trace_assign_type(field, entry); + + trace_seq_printf(s, "[%s] %s:%s:%d\n", + field->correct ? " ok " : " MISS ", + field->func, + field->file, + field->line); + break; + } + case TRACE_USER_STACK: { + struct userstack_entry *field; + + trace_assign_type(field, entry); + + ret = seq_print_userip_objs(field, s, sym_flags); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + ret = trace_seq_putc(s, '\n'); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + break; + } } return TRACE_TYPE_HANDLED; } @@ -1640,6 +2088,7 @@ static enum print_line_t print_raw_fmt(struct trace_iterator *iter) break; } case TRACE_SPECIAL: + case TRACE_USER_STACK: case TRACE_STACK: { struct special_entry *field; @@ -1728,6 +2177,7 @@ static enum print_line_t print_hex_fmt(struct trace_iterator *iter) break; } case TRACE_SPECIAL: + case TRACE_USER_STACK: case TRACE_STACK: { struct special_entry *field; @@ -1782,6 +2232,7 @@ static enum print_line_t print_bin_fmt(struct trace_iterator *iter) break; } case TRACE_SPECIAL: + case TRACE_USER_STACK: case TRACE_STACK: { struct special_entry *field; @@ -1899,6 +2350,11 @@ __tracing_open(struct inode *inode, struct file *file, int *ret) iter->trace = current_trace; iter->pos = -1; + /* Annotate start of buffers if we had overruns */ + if (ring_buffer_overruns(iter->tr->buffer)) + iter->iter_flags |= TRACE_FILE_ANNOTATE; + + for_each_tracing_cpu(cpu) { iter->buffer_iter[cpu] = @@ -1917,10 +2373,7 @@ __tracing_open(struct inode *inode, struct file *file, int *ret) m->private = iter; /* stop the trace while dumping */ - if (iter->tr->ctrl) { - tracer_enabled = 0; - ftrace_function_enabled = 0; - } + tracing_stop(); if (iter->trace && iter->trace->open) iter->trace->open(iter); @@ -1966,14 +2419,7 @@ int tracing_release(struct inode *inode, struct file *file) iter->trace->close(iter); /* reenable tracing if it was previously enabled */ - if (iter->tr->ctrl) { - tracer_enabled = 1; - /* - * It is safe to enable function tracing even if it - * isn't used - */ - ftrace_function_enabled = 1; - } + tracing_start(); mutex_unlock(&trace_types_lock); seq_release(inode, file); @@ -2189,13 +2635,16 @@ static struct file_operations tracing_cpumask_fops = { }; static ssize_t -tracing_iter_ctrl_read(struct file *filp, char __user *ubuf, +tracing_trace_options_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { + int i; char *buf; int r = 0; int len = 0; - int i; + u32 tracer_flags = current_trace->flags->val; + struct tracer_opt *trace_opts = current_trace->flags->opts; + /* calulate max size */ for (i = 0; trace_options[i]; i++) { @@ -2203,6 +2652,15 @@ tracing_iter_ctrl_read(struct file *filp, char __user *ubuf, len += 3; /* "no" and space */ } + /* + * Increase the size with names of options specific + * of the current tracer. + */ + for (i = 0; trace_opts[i].name; i++) { + len += strlen(trace_opts[i].name); + len += 3; /* "no" and space */ + } + /* +2 for \n and \0 */ buf = kmalloc(len + 2, GFP_KERNEL); if (!buf) @@ -2215,6 +2673,15 @@ tracing_iter_ctrl_read(struct file *filp, char __user *ubuf, r += sprintf(buf + r, "no%s ", trace_options[i]); } + for (i = 0; trace_opts[i].name; i++) { + if (tracer_flags & trace_opts[i].bit) + r += sprintf(buf + r, "%s ", + trace_opts[i].name); + else + r += sprintf(buf + r, "no%s ", + trace_opts[i].name); + } + r += sprintf(buf + r, "\n"); WARN_ON(r >= len + 2); @@ -2225,13 +2692,48 @@ tracing_iter_ctrl_read(struct file *filp, char __user *ubuf, return r; } +/* Try to assign a tracer specific option */ +static int set_tracer_option(struct tracer *trace, char *cmp, int neg) +{ + struct tracer_flags *trace_flags = trace->flags; + struct tracer_opt *opts = NULL; + int ret = 0, i = 0; + int len; + + for (i = 0; trace_flags->opts[i].name; i++) { + opts = &trace_flags->opts[i]; + len = strlen(opts->name); + + if (strncmp(cmp, opts->name, len) == 0) { + ret = trace->set_flag(trace_flags->val, + opts->bit, !neg); + break; + } + } + /* Not found */ + if (!trace_flags->opts[i].name) + return -EINVAL; + + /* Refused to handle */ + if (ret) + return ret; + + if (neg) + trace_flags->val &= ~opts->bit; + else + trace_flags->val |= opts->bit; + + return 0; +} + static ssize_t -tracing_iter_ctrl_write(struct file *filp, const char __user *ubuf, +tracing_trace_options_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { char buf[64]; char *cmp = buf; int neg = 0; + int ret; int i; if (cnt >= sizeof(buf)) @@ -2258,11 +2760,13 @@ tracing_iter_ctrl_write(struct file *filp, const char __user *ubuf, break; } } - /* - * If no option could be set, return an error: - */ - if (!trace_options[i]) - return -EINVAL; + + /* If no option could be set, test the specific tracer options */ + if (!trace_options[i]) { + ret = set_tracer_option(current_trace, cmp, neg); + if (ret) + return ret; + } filp->f_pos += cnt; @@ -2271,8 +2775,8 @@ tracing_iter_ctrl_write(struct file *filp, const char __user *ubuf, static struct file_operations tracing_iter_fops = { .open = tracing_open_generic, - .read = tracing_iter_ctrl_read, - .write = tracing_iter_ctrl_write, + .read = tracing_trace_options_read, + .write = tracing_trace_options_write, }; static const char readme_msg[] = @@ -2286,9 +2790,9 @@ static const char readme_msg[] = "# echo sched_switch > /debug/tracing/current_tracer\n" "# cat /debug/tracing/current_tracer\n" "sched_switch\n" - "# cat /debug/tracing/iter_ctrl\n" + "# cat /debug/tracing/trace_options\n" "noprint-parent nosym-offset nosym-addr noverbose\n" - "# echo print-parent > /debug/tracing/iter_ctrl\n" + "# echo print-parent > /debug/tracing/trace_options\n" "# echo 1 > /debug/tracing/tracing_enabled\n" "# cat /debug/tracing/trace > /tmp/trace.txt\n" "echo 0 > /debug/tracing/tracing_enabled\n" @@ -2311,11 +2815,10 @@ static ssize_t tracing_ctrl_read(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { - struct trace_array *tr = filp->private_data; char buf[64]; int r; - r = sprintf(buf, "%ld\n", tr->ctrl); + r = sprintf(buf, "%u\n", tracer_enabled); return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); } @@ -2343,16 +2846,18 @@ tracing_ctrl_write(struct file *filp, const char __user *ubuf, val = !!val; mutex_lock(&trace_types_lock); - if (tr->ctrl ^ val) { - if (val) + if (tracer_enabled ^ val) { + if (val) { tracer_enabled = 1; - else + if (current_trace->start) + current_trace->start(tr); + tracing_start(); + } else { tracer_enabled = 0; - - tr->ctrl = val; - - if (current_trace && current_trace->ctrl_update) - current_trace->ctrl_update(tr); + tracing_stop(); + if (current_trace->stop) + current_trace->stop(tr); + } } mutex_unlock(&trace_types_lock); @@ -2378,29 +2883,11 @@ tracing_set_trace_read(struct file *filp, char __user *ubuf, return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); } -static ssize_t -tracing_set_trace_write(struct file *filp, const char __user *ubuf, - size_t cnt, loff_t *ppos) +static int tracing_set_tracer(char *buf) { struct trace_array *tr = &global_trace; struct tracer *t; - char buf[max_tracer_type_len+1]; - int i; - size_t ret; - - ret = cnt; - - if (cnt > max_tracer_type_len) - cnt = max_tracer_type_len; - - if (copy_from_user(&buf, ubuf, cnt)) - return -EFAULT; - - buf[cnt] = 0; - - /* strip ending whitespace. */ - for (i = cnt - 1; i > 0 && isspace(buf[i]); i--) - buf[i] = 0; + int ret = 0; mutex_lock(&trace_types_lock); for (t = trace_types; t; t = t->next) { @@ -2414,18 +2901,52 @@ tracing_set_trace_write(struct file *filp, const char __user *ubuf, if (t == current_trace) goto out; + trace_branch_disable(); if (current_trace && current_trace->reset) current_trace->reset(tr); current_trace = t; - if (t->init) - t->init(tr); + if (t->init) { + ret = t->init(tr); + if (ret) + goto out; + } + trace_branch_enable(tr); out: mutex_unlock(&trace_types_lock); - if (ret > 0) - filp->f_pos += ret; + return ret; +} + +static ssize_t +tracing_set_trace_write(struct file *filp, const char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + char buf[max_tracer_type_len+1]; + int i; + size_t ret; + int err; + + ret = cnt; + + if (cnt > max_tracer_type_len) + cnt = max_tracer_type_len; + + if (copy_from_user(&buf, ubuf, cnt)) + return -EFAULT; + + buf[cnt] = 0; + + /* strip ending whitespace. */ + for (i = cnt - 1; i > 0 && isspace(buf[i]); i--) + buf[i] = 0; + + err = tracing_set_tracer(buf); + if (err) + return err; + + filp->f_pos += ret; return ret; } @@ -2492,6 +3013,10 @@ static int tracing_open_pipe(struct inode *inode, struct file *filp) return -ENOMEM; mutex_lock(&trace_types_lock); + + /* trace pipe does not show start of buffer */ + cpus_setall(iter->started); + iter->tr = &global_trace; iter->trace = current_trace; filp->private_data = iter; @@ -2667,7 +3192,7 @@ tracing_entries_read(struct file *filp, char __user *ubuf, char buf[64]; int r; - r = sprintf(buf, "%lu\n", tr->entries); + r = sprintf(buf, "%lu\n", tr->entries >> 10); return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); } @@ -2678,7 +3203,6 @@ tracing_entries_write(struct file *filp, const char __user *ubuf, unsigned long val; char buf[64]; int ret, cpu; - struct trace_array *tr = filp->private_data; if (cnt >= sizeof(buf)) return -EINVAL; @@ -2698,12 +3222,7 @@ tracing_entries_write(struct file *filp, const char __user *ubuf, mutex_lock(&trace_types_lock); - if (tr->ctrl) { - cnt = -EBUSY; - pr_info("ftrace: please disable tracing" - " before modifying buffer size\n"); - goto out; - } + tracing_stop(); /* disable all cpu buffers */ for_each_tracing_cpu(cpu) { @@ -2713,6 +3232,9 @@ tracing_entries_write(struct file *filp, const char __user *ubuf, atomic_inc(&max_tr.data[cpu]->disabled); } + /* value is in KB */ + val <<= 10; + if (val != global_trace.entries) { ret = ring_buffer_resize(global_trace.buffer, val); if (ret < 0) { @@ -2751,6 +3273,7 @@ tracing_entries_write(struct file *filp, const char __user *ubuf, atomic_dec(&max_tr.data[cpu]->disabled); } + tracing_start(); max_tr.entries = global_trace.entries; mutex_unlock(&trace_types_lock); @@ -2773,9 +3296,8 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, { char *buf; char *end; - struct trace_array *tr = &global_trace; - if (!tr->ctrl || tracing_disabled) + if (tracing_disabled) return -EINVAL; if (cnt > TRACE_BUF_SIZE) @@ -2841,22 +3363,38 @@ static struct file_operations tracing_mark_fops = { #ifdef CONFIG_DYNAMIC_FTRACE +int __weak ftrace_arch_read_dyn_info(char *buf, int size) +{ + return 0; +} + static ssize_t -tracing_read_long(struct file *filp, char __user *ubuf, +tracing_read_dyn_info(struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos) { + static char ftrace_dyn_info_buffer[1024]; + static DEFINE_MUTEX(dyn_info_mutex); unsigned long *p = filp->private_data; - char buf[64]; + char *buf = ftrace_dyn_info_buffer; + int size = ARRAY_SIZE(ftrace_dyn_info_buffer); int r; - r = sprintf(buf, "%ld\n", *p); + mutex_lock(&dyn_info_mutex); + r = sprintf(buf, "%ld ", *p); - return simple_read_from_buffer(ubuf, cnt, ppos, buf, r); + r += ftrace_arch_read_dyn_info(buf+r, (size-1)-r); + buf[r++] = '\n'; + + r = simple_read_from_buffer(ubuf, cnt, ppos, buf, r); + + mutex_unlock(&dyn_info_mutex); + + return r; } -static struct file_operations tracing_read_long_fops = { +static struct file_operations tracing_dyn_info_fops = { .open = tracing_open_generic, - .read = tracing_read_long, + .read = tracing_read_dyn_info, }; #endif @@ -2897,10 +3435,10 @@ static __init int tracer_init_debugfs(void) if (!entry) pr_warning("Could not create debugfs 'tracing_enabled' entry\n"); - entry = debugfs_create_file("iter_ctrl", 0644, d_tracer, + entry = debugfs_create_file("trace_options", 0644, d_tracer, NULL, &tracing_iter_fops); if (!entry) - pr_warning("Could not create debugfs 'iter_ctrl' entry\n"); + pr_warning("Could not create debugfs 'trace_options' entry\n"); entry = debugfs_create_file("tracing_cpumask", 0644, d_tracer, NULL, &tracing_cpumask_fops); @@ -2950,11 +3488,11 @@ static __init int tracer_init_debugfs(void) pr_warning("Could not create debugfs " "'trace_pipe' entry\n"); - entry = debugfs_create_file("trace_entries", 0644, d_tracer, + entry = debugfs_create_file("buffer_size_kb", 0644, d_tracer, &global_trace, &tracing_entries_fops); if (!entry) pr_warning("Could not create debugfs " - "'trace_entries' entry\n"); + "'buffer_size_kb' entry\n"); entry = debugfs_create_file("trace_marker", 0220, d_tracer, NULL, &tracing_mark_fops); @@ -2965,7 +3503,7 @@ static __init int tracer_init_debugfs(void) #ifdef CONFIG_DYNAMIC_FTRACE entry = debugfs_create_file("dyn_ftrace_total_info", 0444, d_tracer, &ftrace_update_tot_cnt, - &tracing_read_long_fops); + &tracing_dyn_info_fops); if (!entry) pr_warning("Could not create debugfs " "'dyn_ftrace_total_info' entry\n"); @@ -2988,7 +3526,7 @@ int trace_vprintk(unsigned long ip, const char *fmt, va_list args) unsigned long flags, irq_flags; int cpu, len = 0, size, pc; - if (!tr->ctrl || tracing_disabled) + if (tracing_disabled) return 0; pc = preempt_count(); @@ -3046,7 +3584,8 @@ EXPORT_SYMBOL_GPL(__ftrace_printk); static int trace_panic_handler(struct notifier_block *this, unsigned long event, void *unused) { - ftrace_dump(); + if (ftrace_dump_on_oops) + ftrace_dump(); return NOTIFY_OK; } @@ -3062,7 +3601,8 @@ static int trace_die_handler(struct notifier_block *self, { switch (val) { case DIE_OOPS: - ftrace_dump(); + if (ftrace_dump_on_oops) + ftrace_dump(); break; default: break; @@ -3103,7 +3643,6 @@ trace_printk_seq(struct trace_seq *s) trace_seq_reset(s); } - void ftrace_dump(void) { static DEFINE_SPINLOCK(ftrace_dump_lock); @@ -3128,6 +3667,9 @@ void ftrace_dump(void) atomic_inc(&global_trace.data[cpu]->disabled); } + /* don't look at user memory in panic mode */ + trace_flags &= ~TRACE_ITER_SYM_USEROBJ; + printk(KERN_TRACE "Dumping ftrace buffer:\n"); iter.tr = &global_trace; @@ -3221,7 +3763,6 @@ __init static int tracer_alloc_buffers(void) #endif /* All seems OK, enable tracing */ - global_trace.ctrl = tracer_enabled; tracing_disabled = 0; atomic_notifier_chain_register(&panic_notifier_list, diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index 8465ad05270..28c15c2ebc2 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -8,6 +8,7 @@ #include <linux/ring_buffer.h> #include <linux/mmiotrace.h> #include <linux/ftrace.h> +#include <trace/boot.h> enum trace_type { __TRACE_FIRST_TYPE = 0, @@ -21,7 +22,11 @@ enum trace_type { TRACE_SPECIAL, TRACE_MMIO_RW, TRACE_MMIO_MAP, - TRACE_BOOT, + TRACE_BRANCH, + TRACE_BOOT_CALL, + TRACE_BOOT_RET, + TRACE_FN_RET, + TRACE_USER_STACK, __TRACE_LAST_TYPE }; @@ -38,6 +43,7 @@ struct trace_entry { unsigned char flags; unsigned char preempt_count; int pid; + int tgid; }; /* @@ -48,6 +54,16 @@ struct ftrace_entry { unsigned long ip; unsigned long parent_ip; }; + +/* Function return entry */ +struct ftrace_ret_entry { + struct trace_entry ent; + unsigned long ip; + unsigned long parent_ip; + unsigned long long calltime; + unsigned long long rettime; + unsigned long overrun; +}; extern struct tracer boot_tracer; /* @@ -85,6 +101,11 @@ struct stack_entry { unsigned long caller[FTRACE_STACK_ENTRIES]; }; +struct userstack_entry { + struct trace_entry ent; + unsigned long caller[FTRACE_STACK_ENTRIES]; +}; + /* * ftrace_printk entry: */ @@ -112,9 +133,24 @@ struct trace_mmiotrace_map { struct mmiotrace_map map; }; -struct trace_boot { +struct trace_boot_call { struct trace_entry ent; - struct boot_trace initcall; + struct boot_trace_call boot_call; +}; + +struct trace_boot_ret { + struct trace_entry ent; + struct boot_trace_ret boot_ret; +}; + +#define TRACE_FUNC_SIZE 30 +#define TRACE_FILE_SIZE 20 +struct trace_branch { + struct trace_entry ent; + unsigned line; + char func[TRACE_FUNC_SIZE+1]; + char file[TRACE_FILE_SIZE+1]; + char correct; }; /* @@ -172,7 +208,6 @@ struct trace_iterator; struct trace_array { struct ring_buffer *buffer; unsigned long entries; - long ctrl; int cpu; cycle_t time_start; struct task_struct *waiter; @@ -212,13 +247,17 @@ extern void __ftrace_bad_type(void); IF_ASSIGN(var, ent, struct ctx_switch_entry, 0); \ IF_ASSIGN(var, ent, struct trace_field_cont, TRACE_CONT); \ IF_ASSIGN(var, ent, struct stack_entry, TRACE_STACK); \ + IF_ASSIGN(var, ent, struct userstack_entry, TRACE_USER_STACK);\ IF_ASSIGN(var, ent, struct print_entry, TRACE_PRINT); \ IF_ASSIGN(var, ent, struct special_entry, 0); \ IF_ASSIGN(var, ent, struct trace_mmiotrace_rw, \ TRACE_MMIO_RW); \ IF_ASSIGN(var, ent, struct trace_mmiotrace_map, \ TRACE_MMIO_MAP); \ - IF_ASSIGN(var, ent, struct trace_boot, TRACE_BOOT); \ + IF_ASSIGN(var, ent, struct trace_boot_call, TRACE_BOOT_CALL);\ + IF_ASSIGN(var, ent, struct trace_boot_ret, TRACE_BOOT_RET);\ + IF_ASSIGN(var, ent, struct trace_branch, TRACE_BRANCH); \ + IF_ASSIGN(var, ent, struct ftrace_ret_entry, TRACE_FN_RET);\ __ftrace_bad_type(); \ } while (0) @@ -229,29 +268,55 @@ enum print_line_t { TRACE_TYPE_UNHANDLED = 2 /* Relay to other output functions */ }; + +/* + * An option specific to a tracer. This is a boolean value. + * The bit is the bit index that sets its value on the + * flags value in struct tracer_flags. + */ +struct tracer_opt { + const char *name; /* Will appear on the trace_options file */ + u32 bit; /* Mask assigned in val field in tracer_flags */ +}; + +/* + * The set of specific options for a tracer. Your tracer + * have to set the initial value of the flags val. + */ +struct tracer_flags { + u32 val; + struct tracer_opt *opts; +}; + +/* Makes more easy to define a tracer opt */ +#define TRACER_OPT(s, b) .name = #s, .bit = b + /* * A specific tracer, represented by methods that operate on a trace array: */ struct tracer { const char *name; - void (*init)(struct trace_array *tr); + /* Your tracer should raise a warning if init fails */ + int (*init)(struct trace_array *tr); void (*reset)(struct trace_array *tr); + void (*start)(struct trace_array *tr); + void (*stop)(struct trace_array *tr); void (*open)(struct trace_iterator *iter); void (*pipe_open)(struct trace_iterator *iter); void (*close)(struct trace_iterator *iter); - void (*start)(struct trace_iterator *iter); - void (*stop)(struct trace_iterator *iter); ssize_t (*read)(struct trace_iterator *iter, struct file *filp, char __user *ubuf, size_t cnt, loff_t *ppos); - void (*ctrl_update)(struct trace_array *tr); #ifdef CONFIG_FTRACE_STARTUP_TEST int (*selftest)(struct tracer *trace, struct trace_array *tr); #endif enum print_line_t (*print_line)(struct trace_iterator *iter); + /* If you handled the flag setting, return 0 */ + int (*set_flag)(u32 old_flags, u32 bit, int set); struct tracer *next; int print_max; + struct tracer_flags *flags; }; struct trace_seq { @@ -279,8 +344,11 @@ struct trace_iterator { unsigned long iter_flags; loff_t pos; long idx; + + cpumask_t started; }; +int tracing_is_enabled(void); void trace_wake_up(void); void tracing_reset(struct trace_array *tr, int cpu); int tracing_open_generic(struct inode *inode, struct file *filp); @@ -320,9 +388,14 @@ void trace_function(struct trace_array *tr, unsigned long ip, unsigned long parent_ip, unsigned long flags, int pc); +void +trace_function_return(struct ftrace_retfunc *trace); void tracing_start_cmdline_record(void); void tracing_stop_cmdline_record(void); +void tracing_sched_switch_assign_trace(struct trace_array *tr); +void tracing_stop_sched_switch_record(void); +void tracing_start_sched_switch_record(void); int register_tracer(struct tracer *type); void unregister_tracer(struct tracer *type); @@ -383,12 +456,18 @@ extern int trace_selftest_startup_sched_switch(struct tracer *trace, struct trace_array *tr); extern int trace_selftest_startup_sysprof(struct tracer *trace, struct trace_array *tr); +extern int trace_selftest_startup_branch(struct tracer *trace, + struct trace_array *tr); #endif /* CONFIG_FTRACE_STARTUP_TEST */ extern void *head_page(struct trace_array_cpu *data); extern int trace_seq_printf(struct trace_seq *s, const char *fmt, ...); extern void trace_seq_print_cont(struct trace_seq *s, struct trace_iterator *iter); + +extern int +seq_print_ip_sym(struct trace_seq *s, unsigned long ip, + unsigned long sym_flags); extern ssize_t trace_seq_to_user(struct trace_seq *s, char __user *ubuf, size_t cnt); extern long ns2usecs(cycle_t nsec); @@ -396,6 +475,17 @@ extern int trace_vprintk(unsigned long ip, const char *fmt, va_list args); extern unsigned long trace_flags; +/* Standard output formatting function used for function return traces */ +#ifdef CONFIG_FUNCTION_RET_TRACER +extern enum print_line_t print_return_function(struct trace_iterator *iter); +#else +static inline enum print_line_t +print_return_function(struct trace_iterator *iter) +{ + return TRACE_TYPE_UNHANDLED; +} +#endif + /* * trace_iterator_flags is an enumeration that defines bit * positions into trace_flags that controls the output. @@ -415,8 +505,92 @@ enum trace_iterator_flags { TRACE_ITER_STACKTRACE = 0x100, TRACE_ITER_SCHED_TREE = 0x200, TRACE_ITER_PRINTK = 0x400, + TRACE_ITER_PREEMPTONLY = 0x800, + TRACE_ITER_BRANCH = 0x1000, + TRACE_ITER_ANNOTATE = 0x2000, + TRACE_ITER_USERSTACKTRACE = 0x4000, + TRACE_ITER_SYM_USEROBJ = 0x8000 }; +/* + * TRACE_ITER_SYM_MASK masks the options in trace_flags that + * control the output of kernel symbols. + */ +#define TRACE_ITER_SYM_MASK \ + (TRACE_ITER_PRINT_PARENT|TRACE_ITER_SYM_OFFSET|TRACE_ITER_SYM_ADDR) + extern struct tracer nop_trace; +/** + * ftrace_preempt_disable - disable preemption scheduler safe + * + * When tracing can happen inside the scheduler, there exists + * cases that the tracing might happen before the need_resched + * flag is checked. If this happens and the tracer calls + * preempt_enable (after a disable), a schedule might take place + * causing an infinite recursion. + * + * To prevent this, we read the need_recshed flag before + * disabling preemption. When we want to enable preemption we + * check the flag, if it is set, then we call preempt_enable_no_resched. + * Otherwise, we call preempt_enable. + * + * The rational for doing the above is that if need resched is set + * and we have yet to reschedule, we are either in an atomic location + * (where we do not need to check for scheduling) or we are inside + * the scheduler and do not want to resched. + */ +static inline int ftrace_preempt_disable(void) +{ + int resched; + + resched = need_resched(); + preempt_disable_notrace(); + + return resched; +} + +/** + * ftrace_preempt_enable - enable preemption scheduler safe + * @resched: the return value from ftrace_preempt_disable + * + * This is a scheduler safe way to enable preemption and not miss + * any preemption checks. The disabled saved the state of preemption. + * If resched is set, then we were either inside an atomic or + * are inside the scheduler (we would have already scheduled + * otherwise). In this case, we do not want to call normal + * preempt_enable, but preempt_enable_no_resched instead. + */ +static inline void ftrace_preempt_enable(int resched) +{ + if (resched) + preempt_enable_no_resched_notrace(); + else + preempt_enable_notrace(); +} + +#ifdef CONFIG_BRANCH_TRACER +extern int enable_branch_tracing(struct trace_array *tr); +extern void disable_branch_tracing(void); +static inline int trace_branch_enable(struct trace_array *tr) +{ + if (trace_flags & TRACE_ITER_BRANCH) + return enable_branch_tracing(tr); + return 0; +} +static inline void trace_branch_disable(void) +{ + /* due to races, always disable */ + disable_branch_tracing(); +} +#else +static inline int trace_branch_enable(struct trace_array *tr) +{ + return 0; +} +static inline void trace_branch_disable(void) +{ +} +#endif /* CONFIG_BRANCH_TRACER */ + #endif /* _LINUX_KERNEL_TRACE_H */ diff --git a/kernel/trace/trace_boot.c b/kernel/trace/trace_boot.c index d0a5e50eeff..a4fa2c57e34 100644 --- a/kernel/trace/trace_boot.c +++ b/kernel/trace/trace_boot.c @@ -13,73 +13,117 @@ #include "trace.h" static struct trace_array *boot_trace; -static int trace_boot_enabled; +static bool pre_initcalls_finished; - -/* Should be started after do_pre_smp_initcalls() in init/main.c */ +/* Tells the boot tracer that the pre_smp_initcalls are finished. + * So we are ready . + * It doesn't enable sched events tracing however. + * You have to call enable_boot_trace to do so. + */ void start_boot_trace(void) { - trace_boot_enabled = 1; + pre_initcalls_finished = true; } -void stop_boot_trace(void) +void enable_boot_trace(void) { - trace_boot_enabled = 0; + if (pre_initcalls_finished) + tracing_start_sched_switch_record(); } -void reset_boot_trace(struct trace_array *tr) +void disable_boot_trace(void) { - stop_boot_trace(); + if (pre_initcalls_finished) + tracing_stop_sched_switch_record(); } -static void boot_trace_init(struct trace_array *tr) +static void reset_boot_trace(struct trace_array *tr) { int cpu; - boot_trace = tr; - trace_boot_enabled = 0; + tr->time_start = ftrace_now(tr->cpu); + + for_each_online_cpu(cpu) + tracing_reset(tr, cpu); +} + +static int boot_trace_init(struct trace_array *tr) +{ + int cpu; + boot_trace = tr; for_each_cpu_mask(cpu, cpu_possible_map) tracing_reset(tr, cpu); + + tracing_sched_switch_assign_trace(tr); + return 0; } -static void boot_trace_ctrl_update(struct trace_array *tr) +static enum print_line_t +initcall_call_print_line(struct trace_iterator *iter) { - if (tr->ctrl) - start_boot_trace(); + struct trace_entry *entry = iter->ent; + struct trace_seq *s = &iter->seq; + struct trace_boot_call *field; + struct boot_trace_call *call; + u64 ts; + unsigned long nsec_rem; + int ret; + + trace_assign_type(field, entry); + call = &field->boot_call; + ts = iter->ts; + nsec_rem = do_div(ts, 1000000000); + + ret = trace_seq_printf(s, "[%5ld.%09ld] calling %s @ %i\n", + (unsigned long)ts, nsec_rem, call->func, call->caller); + + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; else - stop_boot_trace(); + return TRACE_TYPE_HANDLED; } -static enum print_line_t initcall_print_line(struct trace_iterator *iter) +static enum print_line_t +initcall_ret_print_line(struct trace_iterator *iter) { - int ret; struct trace_entry *entry = iter->ent; - struct trace_boot *field = (struct trace_boot *)entry; - struct boot_trace *it = &field->initcall; struct trace_seq *s = &iter->seq; - struct timespec calltime = ktime_to_timespec(it->calltime); - struct timespec rettime = ktime_to_timespec(it->rettime); - - if (entry->type == TRACE_BOOT) { - ret = trace_seq_printf(s, "[%5ld.%09ld] calling %s @ %i\n", - calltime.tv_sec, - calltime.tv_nsec, - it->func, it->caller); - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; - - ret = trace_seq_printf(s, "[%5ld.%09ld] initcall %s " - "returned %d after %lld msecs\n", - rettime.tv_sec, - rettime.tv_nsec, - it->func, it->result, it->duration); - - if (!ret) - return TRACE_TYPE_PARTIAL_LINE; + struct trace_boot_ret *field; + struct boot_trace_ret *init_ret; + u64 ts; + unsigned long nsec_rem; + int ret; + + trace_assign_type(field, entry); + init_ret = &field->boot_ret; + ts = iter->ts; + nsec_rem = do_div(ts, 1000000000); + + ret = trace_seq_printf(s, "[%5ld.%09ld] initcall %s " + "returned %d after %llu msecs\n", + (unsigned long) ts, + nsec_rem, + init_ret->func, init_ret->result, init_ret->duration); + + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + else return TRACE_TYPE_HANDLED; +} + +static enum print_line_t initcall_print_line(struct trace_iterator *iter) +{ + struct trace_entry *entry = iter->ent; + + switch (entry->type) { + case TRACE_BOOT_CALL: + return initcall_call_print_line(iter); + case TRACE_BOOT_RET: + return initcall_ret_print_line(iter); + default: + return TRACE_TYPE_UNHANDLED; } - return TRACE_TYPE_UNHANDLED; } struct tracer boot_tracer __read_mostly = @@ -87,27 +131,53 @@ struct tracer boot_tracer __read_mostly = .name = "initcall", .init = boot_trace_init, .reset = reset_boot_trace, - .ctrl_update = boot_trace_ctrl_update, .print_line = initcall_print_line, }; -void trace_boot(struct boot_trace *it, initcall_t fn) +void trace_boot_call(struct boot_trace_call *bt, initcall_t fn) { struct ring_buffer_event *event; - struct trace_boot *entry; - struct trace_array_cpu *data; + struct trace_boot_call *entry; unsigned long irq_flags; struct trace_array *tr = boot_trace; - if (!trace_boot_enabled) + if (!pre_initcalls_finished) return; /* Get its name now since this function could * disappear because it is in the .init section. */ - sprint_symbol(it->func, (unsigned long)fn); + sprint_symbol(bt->func, (unsigned long)fn); + preempt_disable(); + + event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), + &irq_flags); + if (!event) + goto out; + entry = ring_buffer_event_data(event); + tracing_generic_entry_update(&entry->ent, 0, 0); + entry->ent.type = TRACE_BOOT_CALL; + entry->boot_call = *bt; + ring_buffer_unlock_commit(tr->buffer, event, irq_flags); + + trace_wake_up(); + + out: + preempt_enable(); +} + +void trace_boot_ret(struct boot_trace_ret *bt, initcall_t fn) +{ + struct ring_buffer_event *event; + struct trace_boot_ret *entry; + unsigned long irq_flags; + struct trace_array *tr = boot_trace; + + if (!pre_initcalls_finished) + return; + + sprint_symbol(bt->func, (unsigned long)fn); preempt_disable(); - data = tr->data[smp_processor_id()]; event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), &irq_flags); @@ -115,8 +185,8 @@ void trace_boot(struct boot_trace *it, initcall_t fn) goto out; entry = ring_buffer_event_data(event); tracing_generic_entry_update(&entry->ent, 0, 0); - entry->ent.type = TRACE_BOOT; - entry->initcall = *it; + entry->ent.type = TRACE_BOOT_RET; + entry->boot_ret = *bt; ring_buffer_unlock_commit(tr->buffer, event, irq_flags); trace_wake_up(); diff --git a/kernel/trace/trace_branch.c b/kernel/trace/trace_branch.c new file mode 100644 index 00000000000..877ee88e6a7 --- /dev/null +++ b/kernel/trace/trace_branch.c @@ -0,0 +1,341 @@ +/* + * unlikely profiler + * + * Copyright (C) 2008 Steven Rostedt <srostedt@redhat.com> + */ +#include <linux/kallsyms.h> +#include <linux/seq_file.h> +#include <linux/spinlock.h> +#include <linux/debugfs.h> +#include <linux/uaccess.h> +#include <linux/module.h> +#include <linux/ftrace.h> +#include <linux/hash.h> +#include <linux/fs.h> +#include <asm/local.h> +#include "trace.h" + +#ifdef CONFIG_BRANCH_TRACER + +static int branch_tracing_enabled __read_mostly; +static DEFINE_MUTEX(branch_tracing_mutex); +static struct trace_array *branch_tracer; + +static void +probe_likely_condition(struct ftrace_branch_data *f, int val, int expect) +{ + struct trace_array *tr = branch_tracer; + struct ring_buffer_event *event; + struct trace_branch *entry; + unsigned long flags, irq_flags; + int cpu, pc; + const char *p; + + /* + * I would love to save just the ftrace_likely_data pointer, but + * this code can also be used by modules. Ugly things can happen + * if the module is unloaded, and then we go and read the + * pointer. This is slower, but much safer. + */ + + if (unlikely(!tr)) + return; + + raw_local_irq_save(flags); + cpu = raw_smp_processor_id(); + if (atomic_inc_return(&tr->data[cpu]->disabled) != 1) + goto out; + + event = ring_buffer_lock_reserve(tr->buffer, sizeof(*entry), + &irq_flags); + if (!event) + goto out; + + pc = preempt_count(); + entry = ring_buffer_event_data(event); + tracing_generic_entry_update(&entry->ent, flags, pc); + entry->ent.type = TRACE_BRANCH; + + /* Strip off the path, only save the file */ + p = f->file + strlen(f->file); + while (p >= f->file && *p != '/') + p--; + p++; + + strncpy(entry->func, f->func, TRACE_FUNC_SIZE); + strncpy(entry->file, p, TRACE_FILE_SIZE); + entry->func[TRACE_FUNC_SIZE] = 0; + entry->file[TRACE_FILE_SIZE] = 0; + entry->line = f->line; + entry->correct = val == expect; + + ring_buffer_unlock_commit(tr->buffer, event, irq_flags); + + out: + atomic_dec(&tr->data[cpu]->disabled); + raw_local_irq_restore(flags); +} + +static inline +void trace_likely_condition(struct ftrace_branch_data *f, int val, int expect) +{ + if (!branch_tracing_enabled) + return; + + probe_likely_condition(f, val, expect); +} + +int enable_branch_tracing(struct trace_array *tr) +{ + int ret = 0; + + mutex_lock(&branch_tracing_mutex); + branch_tracer = tr; + /* + * Must be seen before enabling. The reader is a condition + * where we do not need a matching rmb() + */ + smp_wmb(); + branch_tracing_enabled++; + mutex_unlock(&branch_tracing_mutex); + + return ret; +} + +void disable_branch_tracing(void) +{ + mutex_lock(&branch_tracing_mutex); + + if (!branch_tracing_enabled) + goto out_unlock; + + branch_tracing_enabled--; + + out_unlock: + mutex_unlock(&branch_tracing_mutex); +} + +static void start_branch_trace(struct trace_array *tr) +{ + enable_branch_tracing(tr); +} + +static void stop_branch_trace(struct trace_array *tr) +{ + disable_branch_tracing(); +} + +static int branch_trace_init(struct trace_array *tr) +{ + int cpu; + + for_each_online_cpu(cpu) + tracing_reset(tr, cpu); + + start_branch_trace(tr); + return 0; +} + +static void branch_trace_reset(struct trace_array *tr) +{ + stop_branch_trace(tr); +} + +struct tracer branch_trace __read_mostly = +{ + .name = "branch", + .init = branch_trace_init, + .reset = branch_trace_reset, +#ifdef CONFIG_FTRACE_SELFTEST + .selftest = trace_selftest_startup_branch, +#endif +}; + +__init static int init_branch_trace(void) +{ + return register_tracer(&branch_trace); +} + +device_initcall(init_branch_trace); +#else +static inline +void trace_likely_condition(struct ftrace_branch_data *f, int val, int expect) +{ +} +#endif /* CONFIG_BRANCH_TRACER */ + +void ftrace_likely_update(struct ftrace_branch_data *f, int val, int expect) +{ + /* + * I would love to have a trace point here instead, but the + * trace point code is so inundated with unlikely and likely + * conditions that the recursive nightmare that exists is too + * much to try to get working. At least for now. + */ + trace_likely_condition(f, val, expect); + + /* FIXME: Make this atomic! */ + if (val == expect) + f->correct++; + else + f->incorrect++; +} +EXPORT_SYMBOL(ftrace_likely_update); + +struct ftrace_pointer { + void *start; + void *stop; + int hit; +}; + +static void * +t_next(struct seq_file *m, void *v, loff_t *pos) +{ + const struct ftrace_pointer *f = m->private; + struct ftrace_branch_data *p = v; + + (*pos)++; + + if (v == (void *)1) + return f->start; + + ++p; + + if ((void *)p >= (void *)f->stop) + return NULL; + + return p; +} + +static void *t_start(struct seq_file *m, loff_t *pos) +{ + void *t = (void *)1; + loff_t l = 0; + + for (; t && l < *pos; t = t_next(m, t, &l)) + ; + + return t; +} + +static void t_stop(struct seq_file *m, void *p) +{ +} + +static int t_show(struct seq_file *m, void *v) +{ + const struct ftrace_pointer *fp = m->private; + struct ftrace_branch_data *p = v; + const char *f; + long percent; + + if (v == (void *)1) { + if (fp->hit) + seq_printf(m, " miss hit %% "); + else + seq_printf(m, " correct incorrect %% "); + seq_printf(m, " Function " + " File Line\n" + " ------- --------- - " + " -------- " + " ---- ----\n"); + return 0; + } + + /* Only print the file, not the path */ + f = p->file + strlen(p->file); + while (f >= p->file && *f != '/') + f--; + f++; + + /* + * The miss is overlayed on correct, and hit on incorrect. + */ + if (p->correct) { + percent = p->incorrect * 100; + percent /= p->correct + p->incorrect; + } else + percent = p->incorrect ? 100 : -1; + + seq_printf(m, "%8lu %8lu ", p->correct, p->incorrect); + if (percent < 0) + seq_printf(m, " X "); + else + seq_printf(m, "%3ld ", percent); + seq_printf(m, "%-30.30s %-20.20s %d\n", p->func, f, p->line); + return 0; +} + +static struct seq_operations tracing_likely_seq_ops = { + .start = t_start, + .next = t_next, + .stop = t_stop, + .show = t_show, +}; + +static int tracing_branch_open(struct inode *inode, struct file *file) +{ + int ret; + + ret = seq_open(file, &tracing_likely_seq_ops); + if (!ret) { + struct seq_file *m = file->private_data; + m->private = (void *)inode->i_private; + } + + return ret; +} + +static const struct file_operations tracing_branch_fops = { + .open = tracing_branch_open, + .read = seq_read, + .llseek = seq_lseek, +}; + +#ifdef CONFIG_PROFILE_ALL_BRANCHES +extern unsigned long __start_branch_profile[]; +extern unsigned long __stop_branch_profile[]; + +static const struct ftrace_pointer ftrace_branch_pos = { + .start = __start_branch_profile, + .stop = __stop_branch_profile, + .hit = 1, +}; + +#endif /* CONFIG_PROFILE_ALL_BRANCHES */ + +extern unsigned long __start_annotated_branch_profile[]; +extern unsigned long __stop_annotated_branch_profile[]; + +static const struct ftrace_pointer ftrace_annotated_branch_pos = { + .start = __start_annotated_branch_profile, + .stop = __stop_annotated_branch_profile, +}; + +static __init int ftrace_branch_init(void) +{ + struct dentry *d_tracer; + struct dentry *entry; + + d_tracer = tracing_init_dentry(); + + entry = debugfs_create_file("profile_annotated_branch", 0444, d_tracer, + (void *)&ftrace_annotated_branch_pos, + &tracing_branch_fops); + if (!entry) + pr_warning("Could not create debugfs " + "'profile_annotatet_branch' entry\n"); + +#ifdef CONFIG_PROFILE_ALL_BRANCHES + entry = debugfs_create_file("profile_branch", 0444, d_tracer, + (void *)&ftrace_branch_pos, + &tracing_branch_fops); + if (!entry) + pr_warning("Could not create debugfs" + " 'profile_branch' entry\n"); +#endif + + return 0; +} + +device_initcall(ftrace_branch_init); diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index 0f85a64003d..e74f6d0a321 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -42,24 +42,20 @@ static void stop_function_trace(struct trace_array *tr) tracing_stop_cmdline_record(); } -static void function_trace_init(struct trace_array *tr) +static int function_trace_init(struct trace_array *tr) { - if (tr->ctrl) - start_function_trace(tr); + start_function_trace(tr); + return 0; } static void function_trace_reset(struct trace_array *tr) { - if (tr->ctrl) - stop_function_trace(tr); + stop_function_trace(tr); } -static void function_trace_ctrl_update(struct trace_array *tr) +static void function_trace_start(struct trace_array *tr) { - if (tr->ctrl) - start_function_trace(tr); - else - stop_function_trace(tr); + function_reset(tr); } static struct tracer function_trace __read_mostly = @@ -67,7 +63,7 @@ static struct tracer function_trace __read_mostly = .name = "function", .init = function_trace_init, .reset = function_trace_reset, - .ctrl_update = function_trace_ctrl_update, + .start = function_trace_start, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_function, #endif diff --git a/kernel/trace/trace_functions_return.c b/kernel/trace/trace_functions_return.c new file mode 100644 index 00000000000..e00d64509c9 --- /dev/null +++ b/kernel/trace/trace_functions_return.c @@ -0,0 +1,98 @@ +/* + * + * Function return tracer. + * Copyright (c) 2008 Frederic Weisbecker <fweisbec@gmail.com> + * Mostly borrowed from function tracer which + * is Copyright (c) Steven Rostedt <srostedt@redhat.com> + * + */ +#include <linux/debugfs.h> +#include <linux/uaccess.h> +#include <linux/ftrace.h> +#include <linux/fs.h> + +#include "trace.h" + + +#define TRACE_RETURN_PRINT_OVERRUN 0x1 +static struct tracer_opt trace_opts[] = { + /* Display overruns or not */ + { TRACER_OPT(overrun, TRACE_RETURN_PRINT_OVERRUN) }, + { } /* Empty entry */ +}; + +static struct tracer_flags tracer_flags = { + .val = 0, /* Don't display overruns by default */ + .opts = trace_opts +}; + + +static int return_trace_init(struct trace_array *tr) +{ + int cpu; + for_each_online_cpu(cpu) + tracing_reset(tr, cpu); + + return register_ftrace_return(&trace_function_return); +} + +static void return_trace_reset(struct trace_array *tr) +{ + unregister_ftrace_return(); +} + + +enum print_line_t +print_return_function(struct trace_iterator *iter) +{ + struct trace_seq *s = &iter->seq; + struct trace_entry *entry = iter->ent; + struct ftrace_ret_entry *field; + int ret; + + if (entry->type == TRACE_FN_RET) { + trace_assign_type(field, entry); + ret = trace_seq_printf(s, "%pF -> ", (void *)field->parent_ip); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + ret = seq_print_ip_sym(s, field->ip, + trace_flags & TRACE_ITER_SYM_MASK); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + ret = trace_seq_printf(s, " (%llu ns)", + field->rettime - field->calltime); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + if (tracer_flags.val & TRACE_RETURN_PRINT_OVERRUN) { + ret = trace_seq_printf(s, " (Overruns: %lu)", + field->overrun); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + } + + ret = trace_seq_printf(s, "\n"); + if (!ret) + return TRACE_TYPE_PARTIAL_LINE; + + return TRACE_TYPE_HANDLED; + } + return TRACE_TYPE_UNHANDLED; +} + +static struct tracer return_trace __read_mostly = { + .name = "return", + .init = return_trace_init, + .reset = return_trace_reset, + .print_line = print_return_function, + .flags = &tracer_flags, +}; + +static __init int init_return_trace(void) +{ + return register_tracer(&return_trace); +} + +device_initcall(init_return_trace); diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index 9c74071c10e..7c2e326bbc8 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -353,15 +353,28 @@ void trace_preempt_off(unsigned long a0, unsigned long a1) } #endif /* CONFIG_PREEMPT_TRACER */ +/* + * save_tracer_enabled is used to save the state of the tracer_enabled + * variable when we disable it when we open a trace output file. + */ +static int save_tracer_enabled; + static void start_irqsoff_tracer(struct trace_array *tr) { register_ftrace_function(&trace_ops); - tracer_enabled = 1; + if (tracing_is_enabled()) { + tracer_enabled = 1; + save_tracer_enabled = 1; + } else { + tracer_enabled = 0; + save_tracer_enabled = 0; + } } static void stop_irqsoff_tracer(struct trace_array *tr) { tracer_enabled = 0; + save_tracer_enabled = 0; unregister_ftrace_function(&trace_ops); } @@ -370,53 +383,55 @@ static void __irqsoff_tracer_init(struct trace_array *tr) irqsoff_trace = tr; /* make sure that the tracer is visible */ smp_wmb(); - - if (tr->ctrl) - start_irqsoff_tracer(tr); + start_irqsoff_tracer(tr); } static void irqsoff_tracer_reset(struct trace_array *tr) { - if (tr->ctrl) - stop_irqsoff_tracer(tr); + stop_irqsoff_tracer(tr); } -static void irqsoff_tracer_ctrl_update(struct trace_array *tr) +static void irqsoff_tracer_start(struct trace_array *tr) { - if (tr->ctrl) - start_irqsoff_tracer(tr); - else - stop_irqsoff_tracer(tr); + tracer_enabled = 1; + save_tracer_enabled = 1; +} + +static void irqsoff_tracer_stop(struct trace_array *tr) +{ + tracer_enabled = 0; + save_tracer_enabled = 0; } static void irqsoff_tracer_open(struct trace_iterator *iter) { /* stop the trace while dumping */ - if (iter->tr->ctrl) - stop_irqsoff_tracer(iter->tr); + tracer_enabled = 0; } static void irqsoff_tracer_close(struct trace_iterator *iter) { - if (iter->tr->ctrl) - start_irqsoff_tracer(iter->tr); + /* restart tracing */ + tracer_enabled = save_tracer_enabled; } #ifdef CONFIG_IRQSOFF_TRACER -static void irqsoff_tracer_init(struct trace_array *tr) +static int irqsoff_tracer_init(struct trace_array *tr) { trace_type = TRACER_IRQS_OFF; __irqsoff_tracer_init(tr); + return 0; } static struct tracer irqsoff_tracer __read_mostly = { .name = "irqsoff", .init = irqsoff_tracer_init, .reset = irqsoff_tracer_reset, + .start = irqsoff_tracer_start, + .stop = irqsoff_tracer_stop, .open = irqsoff_tracer_open, .close = irqsoff_tracer_close, - .ctrl_update = irqsoff_tracer_ctrl_update, .print_max = 1, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_irqsoff, @@ -428,11 +443,12 @@ static struct tracer irqsoff_tracer __read_mostly = #endif #ifdef CONFIG_PREEMPT_TRACER -static void preemptoff_tracer_init(struct trace_array *tr) +static int preemptoff_tracer_init(struct trace_array *tr) { trace_type = TRACER_PREEMPT_OFF; __irqsoff_tracer_init(tr); + return 0; } static struct tracer preemptoff_tracer __read_mostly = @@ -440,9 +456,10 @@ static struct tracer preemptoff_tracer __read_mostly = .name = "preemptoff", .init = preemptoff_tracer_init, .reset = irqsoff_tracer_reset, + .start = irqsoff_tracer_start, + .stop = irqsoff_tracer_stop, .open = irqsoff_tracer_open, .close = irqsoff_tracer_close, - .ctrl_update = irqsoff_tracer_ctrl_update, .print_max = 1, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_preemptoff, @@ -456,11 +473,12 @@ static struct tracer preemptoff_tracer __read_mostly = #if defined(CONFIG_IRQSOFF_TRACER) && \ defined(CONFIG_PREEMPT_TRACER) -static void preemptirqsoff_tracer_init(struct trace_array *tr) +static int preemptirqsoff_tracer_init(struct trace_array *tr) { trace_type = TRACER_IRQS_OFF | TRACER_PREEMPT_OFF; __irqsoff_tracer_init(tr); + return 0; } static struct tracer preemptirqsoff_tracer __read_mostly = @@ -468,9 +486,10 @@ static struct tracer preemptirqsoff_tracer __read_mostly = .name = "preemptirqsoff", .init = preemptirqsoff_tracer_init, .reset = irqsoff_tracer_reset, + .start = irqsoff_tracer_start, + .stop = irqsoff_tracer_stop, .open = irqsoff_tracer_open, .close = irqsoff_tracer_close, - .ctrl_update = irqsoff_tracer_ctrl_update, .print_max = 1, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_preemptirqsoff, diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c index e62cbf78eab..2a98a206acc 100644 --- a/kernel/trace/trace_mmiotrace.c +++ b/kernel/trace/trace_mmiotrace.c @@ -32,34 +32,29 @@ static void mmio_reset_data(struct trace_array *tr) tracing_reset(tr, cpu); } -static void mmio_trace_init(struct trace_array *tr) +static int mmio_trace_init(struct trace_array *tr) { pr_debug("in %s\n", __func__); mmio_trace_array = tr; - if (tr->ctrl) { - mmio_reset_data(tr); - enable_mmiotrace(); - } + + mmio_reset_data(tr); + enable_mmiotrace(); + return 0; } static void mmio_trace_reset(struct trace_array *tr) { pr_debug("in %s\n", __func__); - if (tr->ctrl) - disable_mmiotrace(); + + disable_mmiotrace(); mmio_reset_data(tr); mmio_trace_array = NULL; } -static void mmio_trace_ctrl_update(struct trace_array *tr) +static void mmio_trace_start(struct trace_array *tr) { pr_debug("in %s\n", __func__); - if (tr->ctrl) { - mmio_reset_data(tr); - enable_mmiotrace(); - } else { - disable_mmiotrace(); - } + mmio_reset_data(tr); } static int mmio_print_pcidev(struct trace_seq *s, const struct pci_dev *dev) @@ -296,10 +291,10 @@ static struct tracer mmio_tracer __read_mostly = .name = "mmiotrace", .init = mmio_trace_init, .reset = mmio_trace_reset, + .start = mmio_trace_start, .pipe_open = mmio_pipe_open, .close = mmio_close, .read = mmio_read, - .ctrl_update = mmio_trace_ctrl_update, .print_line = mmio_print_line, }; diff --git a/kernel/trace/trace_nop.c b/kernel/trace/trace_nop.c index 4592b486251..b9767acd30a 100644 --- a/kernel/trace/trace_nop.c +++ b/kernel/trace/trace_nop.c @@ -12,6 +12,27 @@ #include "trace.h" +/* Our two options */ +enum { + TRACE_NOP_OPT_ACCEPT = 0x1, + TRACE_NOP_OPT_REFUSE = 0x2 +}; + +/* Options for the tracer (see trace_options file) */ +static struct tracer_opt nop_opts[] = { + /* Option that will be accepted by set_flag callback */ + { TRACER_OPT(test_nop_accept, TRACE_NOP_OPT_ACCEPT) }, + /* Option that will be refused by set_flag callback */ + { TRACER_OPT(test_nop_refuse, TRACE_NOP_OPT_REFUSE) }, + { } /* Always set a last empty entry */ +}; + +static struct tracer_flags nop_flags = { + /* You can check your flags value here when you want. */ + .val = 0, /* By default: all flags disabled */ + .opts = nop_opts +}; + static struct trace_array *ctx_trace; static void start_nop_trace(struct trace_array *tr) @@ -24,7 +45,7 @@ static void stop_nop_trace(struct trace_array *tr) /* Nothing to do! */ } -static void nop_trace_init(struct trace_array *tr) +static int nop_trace_init(struct trace_array *tr) { int cpu; ctx_trace = tr; @@ -32,33 +53,53 @@ static void nop_trace_init(struct trace_array *tr) for_each_online_cpu(cpu) tracing_reset(tr, cpu); - if (tr->ctrl) - start_nop_trace(tr); + start_nop_trace(tr); + return 0; } static void nop_trace_reset(struct trace_array *tr) { - if (tr->ctrl) - stop_nop_trace(tr); + stop_nop_trace(tr); } -static void nop_trace_ctrl_update(struct trace_array *tr) +/* It only serves as a signal handler and a callback to + * accept or refuse tthe setting of a flag. + * If you don't implement it, then the flag setting will be + * automatically accepted. + */ +static int nop_set_flag(u32 old_flags, u32 bit, int set) { - /* When starting a new trace, reset the buffers */ - if (tr->ctrl) - start_nop_trace(tr); - else - stop_nop_trace(tr); + /* + * Note that you don't need to update nop_flags.val yourself. + * The tracing Api will do it automatically if you return 0 + */ + if (bit == TRACE_NOP_OPT_ACCEPT) { + printk(KERN_DEBUG "nop_test_accept flag set to %d: we accept." + " Now cat trace_options to see the result\n", + set); + return 0; + } + + if (bit == TRACE_NOP_OPT_REFUSE) { + printk(KERN_DEBUG "nop_test_refuse flag set to %d: we refuse." + "Now cat trace_options to see the result\n", + set); + return -EINVAL; + } + + return 0; } + struct tracer nop_trace __read_mostly = { .name = "nop", .init = nop_trace_init, .reset = nop_trace_reset, - .ctrl_update = nop_trace_ctrl_update, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_nop, #endif + .flags = &nop_flags, + .set_flag = nop_set_flag }; diff --git a/kernel/trace/trace_sched_switch.c b/kernel/trace/trace_sched_switch.c index b8f56beb1a6..863390557b4 100644 --- a/kernel/trace/trace_sched_switch.c +++ b/kernel/trace/trace_sched_switch.c @@ -16,7 +16,8 @@ static struct trace_array *ctx_trace; static int __read_mostly tracer_enabled; -static atomic_t sched_ref; +static int sched_ref; +static DEFINE_MUTEX(sched_register_mutex); static void probe_sched_switch(struct rq *__rq, struct task_struct *prev, @@ -27,7 +28,7 @@ probe_sched_switch(struct rq *__rq, struct task_struct *prev, int cpu; int pc; - if (!atomic_read(&sched_ref)) + if (!sched_ref) return; tracing_record_cmdline(prev); @@ -123,20 +124,18 @@ static void tracing_sched_unregister(void) static void tracing_start_sched_switch(void) { - long ref; - - ref = atomic_inc_return(&sched_ref); - if (ref == 1) + mutex_lock(&sched_register_mutex); + if (!(sched_ref++)) tracing_sched_register(); + mutex_unlock(&sched_register_mutex); } static void tracing_stop_sched_switch(void) { - long ref; - - ref = atomic_dec_and_test(&sched_ref); - if (ref) + mutex_lock(&sched_register_mutex); + if (!(--sched_ref)) tracing_sched_unregister(); + mutex_unlock(&sched_register_mutex); } void tracing_start_cmdline_record(void) @@ -149,40 +148,86 @@ void tracing_stop_cmdline_record(void) tracing_stop_sched_switch(); } +/** + * tracing_start_sched_switch_record - start tracing context switches + * + * Turns on context switch tracing for a tracer. + */ +void tracing_start_sched_switch_record(void) +{ + if (unlikely(!ctx_trace)) { + WARN_ON(1); + return; + } + + tracing_start_sched_switch(); + + mutex_lock(&sched_register_mutex); + tracer_enabled++; + mutex_unlock(&sched_register_mutex); +} + +/** + * tracing_stop_sched_switch_record - start tracing context switches + * + * Turns off context switch tracing for a tracer. + */ +void tracing_stop_sched_switch_record(void) +{ + mutex_lock(&sched_register_mutex); + tracer_enabled--; + WARN_ON(tracer_enabled < 0); + mutex_unlock(&sched_register_mutex); + + tracing_stop_sched_switch(); +} + +/** + * tracing_sched_switch_assign_trace - assign a trace array for ctx switch + * @tr: trace array pointer to assign + * + * Some tracers might want to record the context switches in their + * trace. This function lets those tracers assign the trace array + * to use. + */ +void tracing_sched_switch_assign_trace(struct trace_array *tr) +{ + ctx_trace = tr; +} + static void start_sched_trace(struct trace_array *tr) { sched_switch_reset(tr); - tracing_start_cmdline_record(); - tracer_enabled = 1; + tracing_start_sched_switch_record(); } static void stop_sched_trace(struct trace_array *tr) { - tracer_enabled = 0; - tracing_stop_cmdline_record(); + tracing_stop_sched_switch_record(); } -static void sched_switch_trace_init(struct trace_array *tr) +static int sched_switch_trace_init(struct trace_array *tr) { ctx_trace = tr; - - if (tr->ctrl) - start_sched_trace(tr); + start_sched_trace(tr); + return 0; } static void sched_switch_trace_reset(struct trace_array *tr) { - if (tr->ctrl) + if (sched_ref) stop_sched_trace(tr); } -static void sched_switch_trace_ctrl_update(struct trace_array *tr) +static void sched_switch_trace_start(struct trace_array *tr) { - /* When starting a new trace, reset the buffers */ - if (tr->ctrl) - start_sched_trace(tr); - else - stop_sched_trace(tr); + sched_switch_reset(tr); + tracing_start_sched_switch(); +} + +static void sched_switch_trace_stop(struct trace_array *tr) +{ + tracing_stop_sched_switch(); } static struct tracer sched_switch_trace __read_mostly = @@ -190,7 +235,8 @@ static struct tracer sched_switch_trace __read_mostly = .name = "sched_switch", .init = sched_switch_trace_init, .reset = sched_switch_trace_reset, - .ctrl_update = sched_switch_trace_ctrl_update, + .start = sched_switch_trace_start, + .stop = sched_switch_trace_stop, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_sched_switch, #endif @@ -198,14 +244,6 @@ static struct tracer sched_switch_trace __read_mostly = __init static int init_sched_switch_trace(void) { - int ret = 0; - - if (atomic_read(&sched_ref)) - ret = tracing_sched_register(); - if (ret) { - pr_info("error registering scheduler trace\n"); - return ret; - } return register_tracer(&sched_switch_trace); } device_initcall(init_sched_switch_trace); diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 3ae93f16b56..0067b49746c 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -50,8 +50,7 @@ wakeup_tracer_call(unsigned long ip, unsigned long parent_ip) return; pc = preempt_count(); - resched = need_resched(); - preempt_disable_notrace(); + resched = ftrace_preempt_disable(); cpu = raw_smp_processor_id(); data = tr->data[cpu]; @@ -81,15 +80,7 @@ wakeup_tracer_call(unsigned long ip, unsigned long parent_ip) out: atomic_dec(&data->disabled); - /* - * To prevent recursion from the scheduler, if the - * resched flag was set before we entered, then - * don't reschedule. - */ - if (resched) - preempt_enable_no_resched_notrace(); - else - preempt_enable_notrace(); + ftrace_preempt_enable(resched); } static struct ftrace_ops trace_ops __read_mostly = @@ -271,6 +262,12 @@ out: atomic_dec(&wakeup_trace->data[cpu]->disabled); } +/* + * save_tracer_enabled is used to save the state of the tracer_enabled + * variable when we disable it when we open a trace output file. + */ +static int save_tracer_enabled; + static void start_wakeup_tracer(struct trace_array *tr) { int ret; @@ -309,7 +306,13 @@ static void start_wakeup_tracer(struct trace_array *tr) register_ftrace_function(&trace_ops); - tracer_enabled = 1; + if (tracing_is_enabled()) { + tracer_enabled = 1; + save_tracer_enabled = 1; + } else { + tracer_enabled = 0; + save_tracer_enabled = 0; + } return; fail_deprobe_wake_new: @@ -321,49 +324,53 @@ fail_deprobe: static void stop_wakeup_tracer(struct trace_array *tr) { tracer_enabled = 0; + save_tracer_enabled = 0; unregister_ftrace_function(&trace_ops); unregister_trace_sched_switch(probe_wakeup_sched_switch); unregister_trace_sched_wakeup_new(probe_wakeup); unregister_trace_sched_wakeup(probe_wakeup); } -static void wakeup_tracer_init(struct trace_array *tr) +static int wakeup_tracer_init(struct trace_array *tr) { wakeup_trace = tr; - - if (tr->ctrl) - start_wakeup_tracer(tr); + start_wakeup_tracer(tr); + return 0; } static void wakeup_tracer_reset(struct trace_array *tr) { - if (tr->ctrl) { - stop_wakeup_tracer(tr); - /* make sure we put back any tasks we are tracing */ - wakeup_reset(tr); - } + stop_wakeup_tracer(tr); + /* make sure we put back any tasks we are tracing */ + wakeup_reset(tr); +} + +static void wakeup_tracer_start(struct trace_array *tr) +{ + wakeup_reset(tr); + tracer_enabled = 1; + save_tracer_enabled = 1; } -static void wakeup_tracer_ctrl_update(struct trace_array *tr) +static void wakeup_tracer_stop(struct trace_array *tr) { - if (tr->ctrl) - start_wakeup_tracer(tr); - else - stop_wakeup_tracer(tr); + tracer_enabled = 0; + save_tracer_enabled = 0; } static void wakeup_tracer_open(struct trace_iterator *iter) { /* stop the trace while dumping */ - if (iter->tr->ctrl) - stop_wakeup_tracer(iter->tr); + tracer_enabled = 0; } static void wakeup_tracer_close(struct trace_iterator *iter) { /* forget about any processes we were recording */ - if (iter->tr->ctrl) - start_wakeup_tracer(iter->tr); + if (save_tracer_enabled) { + wakeup_reset(iter->tr); + tracer_enabled = 1; + } } static struct tracer wakeup_tracer __read_mostly = @@ -371,9 +378,10 @@ static struct tracer wakeup_tracer __read_mostly = .name = "wakeup", .init = wakeup_tracer_init, .reset = wakeup_tracer_reset, + .start = wakeup_tracer_start, + .stop = wakeup_tracer_stop, .open = wakeup_tracer_open, .close = wakeup_tracer_close, - .ctrl_update = wakeup_tracer_ctrl_update, .print_max = 1, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_wakeup, diff --git a/kernel/trace/trace_selftest.c b/kernel/trace/trace_selftest.c index 90bc752a758..88c8eb70f54 100644 --- a/kernel/trace/trace_selftest.c +++ b/kernel/trace/trace_selftest.c @@ -13,6 +13,7 @@ static inline int trace_valid_entry(struct trace_entry *entry) case TRACE_STACK: case TRACE_PRINT: case TRACE_SPECIAL: + case TRACE_BRANCH: return 1; } return 0; @@ -51,7 +52,7 @@ static int trace_test_buffer(struct trace_array *tr, unsigned long *count) int cpu, ret = 0; /* Don't allow flipping of max traces now */ - raw_local_irq_save(flags); + local_irq_save(flags); __raw_spin_lock(&ftrace_max_lock); cnt = ring_buffer_entries(tr->buffer); @@ -62,7 +63,7 @@ static int trace_test_buffer(struct trace_array *tr, unsigned long *count) break; } __raw_spin_unlock(&ftrace_max_lock); - raw_local_irq_restore(flags); + local_irq_restore(flags); if (count) *count = cnt; @@ -70,6 +71,11 @@ static int trace_test_buffer(struct trace_array *tr, unsigned long *count) return ret; } +static inline void warn_failed_init_tracer(struct tracer *trace, int init_ret) +{ + printk(KERN_WARNING "Failed to init %s tracer, init returned %d\n", + trace->name, init_ret); +} #ifdef CONFIG_FUNCTION_TRACER #ifdef CONFIG_DYNAMIC_FTRACE @@ -110,8 +116,11 @@ int trace_selftest_startup_dynamic_tracing(struct tracer *trace, ftrace_set_filter(func_name, strlen(func_name), 1); /* enable tracing */ - tr->ctrl = 1; - trace->init(tr); + ret = trace->init(tr); + if (ret) { + warn_failed_init_tracer(trace, ret); + goto out; + } /* Sleep for a 1/10 of a second */ msleep(100); @@ -134,13 +143,13 @@ int trace_selftest_startup_dynamic_tracing(struct tracer *trace, msleep(100); /* stop the tracing. */ - tr->ctrl = 0; - trace->ctrl_update(tr); + tracing_stop(); ftrace_enabled = 0; /* check the trace buffer */ ret = trace_test_buffer(tr, &count); trace->reset(tr); + tracing_start(); /* we should only have one item */ if (!ret && count != 1) { @@ -148,6 +157,7 @@ int trace_selftest_startup_dynamic_tracing(struct tracer *trace, ret = -1; goto out; } + out: ftrace_enabled = save_ftrace_enabled; tracer_enabled = save_tracer_enabled; @@ -180,18 +190,22 @@ trace_selftest_startup_function(struct tracer *trace, struct trace_array *tr) ftrace_enabled = 1; tracer_enabled = 1; - tr->ctrl = 1; - trace->init(tr); + ret = trace->init(tr); + if (ret) { + warn_failed_init_tracer(trace, ret); + goto out; + } + /* Sleep for a 1/10 of a second */ msleep(100); /* stop the tracing. */ - tr->ctrl = 0; - trace->ctrl_update(tr); + tracing_stop(); ftrace_enabled = 0; /* check the trace buffer */ ret = trace_test_buffer(tr, &count); trace->reset(tr); + tracing_start(); if (!ret && !count) { printk(KERN_CONT ".. no entries found .."); @@ -223,8 +237,12 @@ trace_selftest_startup_irqsoff(struct tracer *trace, struct trace_array *tr) int ret; /* start the tracing */ - tr->ctrl = 1; - trace->init(tr); + ret = trace->init(tr); + if (ret) { + warn_failed_init_tracer(trace, ret); + return ret; + } + /* reset the max latency */ tracing_max_latency = 0; /* disable interrupts for a bit */ @@ -232,13 +250,13 @@ trace_selftest_startup_irqsoff(struct tracer *trace, struct trace_array *tr) udelay(100); local_irq_enable(); /* stop the tracing. */ - tr->ctrl = 0; - trace->ctrl_update(tr); + tracing_stop(); /* check both trace buffers */ ret = trace_test_buffer(tr, NULL); if (!ret) ret = trace_test_buffer(&max_tr, &count); trace->reset(tr); + tracing_start(); if (!ret && !count) { printk(KERN_CONT ".. no entries found .."); @@ -259,9 +277,26 @@ trace_selftest_startup_preemptoff(struct tracer *trace, struct trace_array *tr) unsigned long count; int ret; + /* + * Now that the big kernel lock is no longer preemptable, + * and this is called with the BKL held, it will always + * fail. If preemption is already disabled, simply + * pass the test. When the BKL is removed, or becomes + * preemptible again, we will once again test this, + * so keep it in. + */ + if (preempt_count()) { + printk(KERN_CONT "can not test ... force "); + return 0; + } + /* start the tracing */ - tr->ctrl = 1; - trace->init(tr); + ret = trace->init(tr); + if (ret) { + warn_failed_init_tracer(trace, ret); + return ret; + } + /* reset the max latency */ tracing_max_latency = 0; /* disable preemption for a bit */ @@ -269,13 +304,13 @@ trace_selftest_startup_preemptoff(struct tracer *trace, struct trace_array *tr) udelay(100); preempt_enable(); /* stop the tracing. */ - tr->ctrl = 0; - trace->ctrl_update(tr); + tracing_stop(); /* check both trace buffers */ ret = trace_test_buffer(tr, NULL); if (!ret) ret = trace_test_buffer(&max_tr, &count); trace->reset(tr); + tracing_start(); if (!ret && !count) { printk(KERN_CONT ".. no entries found .."); @@ -296,9 +331,25 @@ trace_selftest_startup_preemptirqsoff(struct tracer *trace, struct trace_array * unsigned long count; int ret; + /* + * Now that the big kernel lock is no longer preemptable, + * and this is called with the BKL held, it will always + * fail. If preemption is already disabled, simply + * pass the test. When the BKL is removed, or becomes + * preemptible again, we will once again test this, + * so keep it in. + */ + if (preempt_count()) { + printk(KERN_CONT "can not test ... force "); + return 0; + } + /* start the tracing */ - tr->ctrl = 1; - trace->init(tr); + ret = trace->init(tr); + if (ret) { + warn_failed_init_tracer(trace, ret); + goto out; + } /* reset the max latency */ tracing_max_latency = 0; @@ -312,27 +363,30 @@ trace_selftest_startup_preemptirqsoff(struct tracer *trace, struct trace_array * local_irq_enable(); /* stop the tracing. */ - tr->ctrl = 0; - trace->ctrl_update(tr); + tracing_stop(); /* check both trace buffers */ ret = trace_test_buffer(tr, NULL); - if (ret) + if (ret) { + tracing_start(); goto out; + } ret = trace_test_buffer(&max_tr, &count); - if (ret) + if (ret) { + tracing_start(); goto out; + } if (!ret && !count) { printk(KERN_CONT ".. no entries found .."); ret = -1; + tracing_start(); goto out; } /* do the test by disabling interrupts first this time */ tracing_max_latency = 0; - tr->ctrl = 1; - trace->ctrl_update(tr); + tracing_start(); preempt_disable(); local_irq_disable(); udelay(100); @@ -341,8 +395,7 @@ trace_selftest_startup_preemptirqsoff(struct tracer *trace, struct trace_array * local_irq_enable(); /* stop the tracing. */ - tr->ctrl = 0; - trace->ctrl_update(tr); + tracing_stop(); /* check both trace buffers */ ret = trace_test_buffer(tr, NULL); if (ret) @@ -358,6 +411,7 @@ trace_selftest_startup_preemptirqsoff(struct tracer *trace, struct trace_array * out: trace->reset(tr); + tracing_start(); tracing_max_latency = save_max; return ret; @@ -423,8 +477,12 @@ trace_selftest_startup_wakeup(struct tracer *trace, struct trace_array *tr) wait_for_completion(&isrt); /* start the tracing */ - tr->ctrl = 1; - trace->init(tr); + ret = trace->init(tr); + if (ret) { + warn_failed_init_tracer(trace, ret); + return ret; + } + /* reset the max latency */ tracing_max_latency = 0; @@ -448,8 +506,7 @@ trace_selftest_startup_wakeup(struct tracer *trace, struct trace_array *tr) msleep(100); /* stop the tracing. */ - tr->ctrl = 0; - trace->ctrl_update(tr); + tracing_stop(); /* check both trace buffers */ ret = trace_test_buffer(tr, NULL); if (!ret) @@ -457,6 +514,7 @@ trace_selftest_startup_wakeup(struct tracer *trace, struct trace_array *tr) trace->reset(tr); + tracing_start(); tracing_max_latency = save_max; @@ -480,16 +538,20 @@ trace_selftest_startup_sched_switch(struct tracer *trace, struct trace_array *tr int ret; /* start the tracing */ - tr->ctrl = 1; - trace->init(tr); + ret = trace->init(tr); + if (ret) { + warn_failed_init_tracer(trace, ret); + return ret; + } + /* Sleep for a 1/10 of a second */ msleep(100); /* stop the tracing. */ - tr->ctrl = 0; - trace->ctrl_update(tr); + tracing_stop(); /* check the trace buffer */ ret = trace_test_buffer(tr, &count); trace->reset(tr); + tracing_start(); if (!ret && !count) { printk(KERN_CONT ".. no entries found .."); @@ -508,17 +570,48 @@ trace_selftest_startup_sysprof(struct tracer *trace, struct trace_array *tr) int ret; /* start the tracing */ - tr->ctrl = 1; - trace->init(tr); + ret = trace->init(tr); + if (ret) { + warn_failed_init_tracer(trace, ret); + return 0; + } + /* Sleep for a 1/10 of a second */ msleep(100); /* stop the tracing. */ - tr->ctrl = 0; - trace->ctrl_update(tr); + tracing_stop(); /* check the trace buffer */ ret = trace_test_buffer(tr, &count); trace->reset(tr); + tracing_start(); return ret; } #endif /* CONFIG_SYSPROF_TRACER */ + +#ifdef CONFIG_BRANCH_TRACER +int +trace_selftest_startup_branch(struct tracer *trace, struct trace_array *tr) +{ + unsigned long count; + int ret; + + /* start the tracing */ + ret = trace->init(tr); + if (ret) { + warn_failed_init_tracer(trace, ret); + return ret; + } + + /* Sleep for a 1/10 of a second */ + msleep(100); + /* stop the tracing. */ + tracing_stop(); + /* check the trace buffer */ + ret = trace_test_buffer(tr, &count); + trace->reset(tr); + tracing_start(); + + return ret; +} +#endif /* CONFIG_BRANCH_TRACER */ diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c index 3bdb44bde4b..fde3be15c64 100644 --- a/kernel/trace/trace_stack.c +++ b/kernel/trace/trace_stack.c @@ -107,8 +107,7 @@ stack_trace_call(unsigned long ip, unsigned long parent_ip) if (unlikely(!ftrace_enabled || stack_trace_disabled)) return; - resched = need_resched(); - preempt_disable_notrace(); + resched = ftrace_preempt_disable(); cpu = raw_smp_processor_id(); /* no atomic needed, we only modify this variable by this cpu */ @@ -120,10 +119,7 @@ stack_trace_call(unsigned long ip, unsigned long parent_ip) out: per_cpu(trace_active, cpu)--; /* prevent recursion in schedule */ - if (resched) - preempt_enable_no_resched_notrace(); - else - preempt_enable_notrace(); + ftrace_preempt_enable(resched); } static struct ftrace_ops trace_ops __read_mostly = diff --git a/kernel/trace/trace_sysprof.c b/kernel/trace/trace_sysprof.c index 9587d3bcba5..54960edb96d 100644 --- a/kernel/trace/trace_sysprof.c +++ b/kernel/trace/trace_sysprof.c @@ -261,27 +261,17 @@ static void stop_stack_trace(struct trace_array *tr) mutex_unlock(&sample_timer_lock); } -static void stack_trace_init(struct trace_array *tr) +static int stack_trace_init(struct trace_array *tr) { sysprof_trace = tr; - if (tr->ctrl) - start_stack_trace(tr); + start_stack_trace(tr); + return 0; } static void stack_trace_reset(struct trace_array *tr) { - if (tr->ctrl) - stop_stack_trace(tr); -} - -static void stack_trace_ctrl_update(struct trace_array *tr) -{ - /* When starting a new trace, reset the buffers */ - if (tr->ctrl) - start_stack_trace(tr); - else - stop_stack_trace(tr); + stop_stack_trace(tr); } static struct tracer stack_trace __read_mostly = @@ -289,7 +279,6 @@ static struct tracer stack_trace __read_mostly = .name = "sysprof", .init = stack_trace_init, .reset = stack_trace_reset, - .ctrl_update = stack_trace_ctrl_update, #ifdef CONFIG_FTRACE_SELFTEST .selftest = trace_selftest_startup_sysprof, #endif diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index af8c8566488..79602740bbb 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -43,6 +43,7 @@ static DEFINE_MUTEX(tracepoints_mutex); */ #define TRACEPOINT_HASH_BITS 6 #define TRACEPOINT_TABLE_SIZE (1 << TRACEPOINT_HASH_BITS) +static struct hlist_head tracepoint_table[TRACEPOINT_TABLE_SIZE]; /* * Note about RCU : @@ -54,40 +55,43 @@ struct tracepoint_entry { struct hlist_node hlist; void **funcs; int refcount; /* Number of times armed. 0 if disarmed. */ - struct rcu_head rcu; - void *oldptr; - unsigned char rcu_pending:1; char name[0]; }; -static struct hlist_head tracepoint_table[TRACEPOINT_TABLE_SIZE]; +struct tp_probes { + union { + struct rcu_head rcu; + struct list_head list; + } u; + void *probes[0]; +}; -static void free_old_closure(struct rcu_head *head) +static inline void *allocate_probes(int count) { - struct tracepoint_entry *entry = container_of(head, - struct tracepoint_entry, rcu); - kfree(entry->oldptr); - /* Make sure we free the data before setting the pending flag to 0 */ - smp_wmb(); - entry->rcu_pending = 0; + struct tp_probes *p = kmalloc(count * sizeof(void *) + + sizeof(struct tp_probes), GFP_KERNEL); + return p == NULL ? NULL : p->probes; } -static void tracepoint_entry_free_old(struct tracepoint_entry *entry, void *old) +static void rcu_free_old_probes(struct rcu_head *head) { - if (!old) - return; - entry->oldptr = old; - entry->rcu_pending = 1; - /* write rcu_pending before calling the RCU callback */ - smp_wmb(); - call_rcu_sched(&entry->rcu, free_old_closure); + kfree(container_of(head, struct tp_probes, u.rcu)); +} + +static inline void release_probes(void *old) +{ + if (old) { + struct tp_probes *tp_probes = container_of(old, + struct tp_probes, probes[0]); + call_rcu_sched(&tp_probes->u.rcu, rcu_free_old_probes); + } } static void debug_print_probes(struct tracepoint_entry *entry) { int i; - if (!tracepoint_debug) + if (!tracepoint_debug || !entry->funcs) return; for (i = 0; entry->funcs[i]; i++) @@ -111,12 +115,13 @@ tracepoint_entry_add_probe(struct tracepoint_entry *entry, void *probe) return ERR_PTR(-EEXIST); } /* + 2 : one for new probe, one for NULL func */ - new = kzalloc((nr_probes + 2) * sizeof(void *), GFP_KERNEL); + new = allocate_probes(nr_probes + 2); if (new == NULL) return ERR_PTR(-ENOMEM); if (old) memcpy(new, old, nr_probes * sizeof(void *)); new[nr_probes] = probe; + new[nr_probes + 1] = NULL; entry->refcount = nr_probes + 1; entry->funcs = new; debug_print_probes(entry); @@ -132,7 +137,7 @@ tracepoint_entry_remove_probe(struct tracepoint_entry *entry, void *probe) old = entry->funcs; if (!old) - return NULL; + return ERR_PTR(-ENOENT); debug_print_probes(entry); /* (N -> M), (N > 1, M >= 0) probes */ @@ -151,13 +156,13 @@ tracepoint_entry_remove_probe(struct tracepoint_entry *entry, void *probe) int j = 0; /* N -> M, (N > 1, M > 0) */ /* + 1 for NULL */ - new = kzalloc((nr_probes - nr_del + 1) - * sizeof(void *), GFP_KERNEL); + new = allocate_probes(nr_probes - nr_del + 1); if (new == NULL) return ERR_PTR(-ENOMEM); for (i = 0; old[i]; i++) if ((probe && old[i] != probe)) new[j++] = old[i]; + new[nr_probes - nr_del] = NULL; entry->refcount = nr_probes - nr_del; entry->funcs = new; } @@ -215,7 +220,6 @@ static struct tracepoint_entry *add_tracepoint(const char *name) memcpy(&e->name[0], name, name_len); e->funcs = NULL; e->refcount = 0; - e->rcu_pending = 0; hlist_add_head(&e->hlist, head); return e; } @@ -224,32 +228,10 @@ static struct tracepoint_entry *add_tracepoint(const char *name) * Remove the tracepoint from the tracepoint hash table. Must be called with * mutex_lock held. */ -static int remove_tracepoint(const char *name) +static inline void remove_tracepoint(struct tracepoint_entry *e) { - struct hlist_head *head; - struct hlist_node *node; - struct tracepoint_entry *e; - int found = 0; - size_t len = strlen(name) + 1; - u32 hash = jhash(name, len-1, 0); - - head = &tracepoint_table[hash & (TRACEPOINT_TABLE_SIZE - 1)]; - hlist_for_each_entry(e, node, head, hlist) { - if (!strcmp(name, e->name)) { - found = 1; - break; - } - } - if (!found) - return -ENOENT; - if (e->refcount) - return -EBUSY; hlist_del(&e->hlist); - /* Make sure the call_rcu_sched has been executed */ - if (e->rcu_pending) - rcu_barrier_sched(); kfree(e); - return 0; } /* @@ -280,6 +262,7 @@ static void set_tracepoint(struct tracepoint_entry **entry, static void disable_tracepoint(struct tracepoint *elem) { elem->state = 0; + rcu_assign_pointer(elem->funcs, NULL); } /** @@ -320,6 +303,23 @@ static void tracepoint_update_probes(void) module_update_tracepoints(); } +static void *tracepoint_add_probe(const char *name, void *probe) +{ + struct tracepoint_entry *entry; + void *old; + + entry = get_tracepoint(name); + if (!entry) { + entry = add_tracepoint(name); + if (IS_ERR(entry)) + return entry; + } + old = tracepoint_entry_add_probe(entry, probe); + if (IS_ERR(old) && !entry->refcount) + remove_tracepoint(entry); + return old; +} + /** * tracepoint_probe_register - Connect a probe to a tracepoint * @name: tracepoint name @@ -330,44 +330,36 @@ static void tracepoint_update_probes(void) */ int tracepoint_probe_register(const char *name, void *probe) { - struct tracepoint_entry *entry; - int ret = 0; void *old; mutex_lock(&tracepoints_mutex); - entry = get_tracepoint(name); - if (!entry) { - entry = add_tracepoint(name); - if (IS_ERR(entry)) { - ret = PTR_ERR(entry); - goto end; - } - } - /* - * If we detect that a call_rcu_sched is pending for this tracepoint, - * make sure it's executed now. - */ - if (entry->rcu_pending) - rcu_barrier_sched(); - old = tracepoint_entry_add_probe(entry, probe); - if (IS_ERR(old)) { - ret = PTR_ERR(old); - goto end; - } + old = tracepoint_add_probe(name, probe); mutex_unlock(&tracepoints_mutex); + if (IS_ERR(old)) + return PTR_ERR(old); + tracepoint_update_probes(); /* may update entry */ - mutex_lock(&tracepoints_mutex); - entry = get_tracepoint(name); - WARN_ON(!entry); - if (entry->rcu_pending) - rcu_barrier_sched(); - tracepoint_entry_free_old(entry, old); -end: - mutex_unlock(&tracepoints_mutex); - return ret; + release_probes(old); + return 0; } EXPORT_SYMBOL_GPL(tracepoint_probe_register); +static void *tracepoint_remove_probe(const char *name, void *probe) +{ + struct tracepoint_entry *entry; + void *old; + + entry = get_tracepoint(name); + if (!entry) + return ERR_PTR(-ENOENT); + old = tracepoint_entry_remove_probe(entry, probe); + if (IS_ERR(old)) + return old; + if (!entry->refcount) + remove_tracepoint(entry); + return old; +} + /** * tracepoint_probe_unregister - Disconnect a probe from a tracepoint * @name: tracepoint name @@ -380,38 +372,104 @@ EXPORT_SYMBOL_GPL(tracepoint_probe_register); */ int tracepoint_probe_unregister(const char *name, void *probe) { - struct tracepoint_entry *entry; void *old; - int ret = -ENOENT; mutex_lock(&tracepoints_mutex); - entry = get_tracepoint(name); - if (!entry) - goto end; - if (entry->rcu_pending) - rcu_barrier_sched(); - old = tracepoint_entry_remove_probe(entry, probe); - if (!old) { - printk(KERN_WARNING "Warning: Trying to unregister a probe" - "that doesn't exist\n"); - goto end; - } + old = tracepoint_remove_probe(name, probe); mutex_unlock(&tracepoints_mutex); + if (IS_ERR(old)) + return PTR_ERR(old); + tracepoint_update_probes(); /* may update entry */ + release_probes(old); + return 0; +} +EXPORT_SYMBOL_GPL(tracepoint_probe_unregister); + +static LIST_HEAD(old_probes); +static int need_update; + +static void tracepoint_add_old_probes(void *old) +{ + need_update = 1; + if (old) { + struct tp_probes *tp_probes = container_of(old, + struct tp_probes, probes[0]); + list_add(&tp_probes->u.list, &old_probes); + } +} + +/** + * tracepoint_probe_register_noupdate - register a probe but not connect + * @name: tracepoint name + * @probe: probe handler + * + * caller must call tracepoint_probe_update_all() + */ +int tracepoint_probe_register_noupdate(const char *name, void *probe) +{ + void *old; + mutex_lock(&tracepoints_mutex); - entry = get_tracepoint(name); - if (!entry) - goto end; - if (entry->rcu_pending) - rcu_barrier_sched(); - tracepoint_entry_free_old(entry, old); - remove_tracepoint(name); /* Ignore busy error message */ - ret = 0; -end: + old = tracepoint_add_probe(name, probe); + if (IS_ERR(old)) { + mutex_unlock(&tracepoints_mutex); + return PTR_ERR(old); + } + tracepoint_add_old_probes(old); mutex_unlock(&tracepoints_mutex); - return ret; + return 0; } -EXPORT_SYMBOL_GPL(tracepoint_probe_unregister); +EXPORT_SYMBOL_GPL(tracepoint_probe_register_noupdate); + +/** + * tracepoint_probe_unregister_noupdate - remove a probe but not disconnect + * @name: tracepoint name + * @probe: probe function pointer + * + * caller must call tracepoint_probe_update_all() + */ +int tracepoint_probe_unregister_noupdate(const char *name, void *probe) +{ + void *old; + + mutex_lock(&tracepoints_mutex); + old = tracepoint_remove_probe(name, probe); + if (IS_ERR(old)) { + mutex_unlock(&tracepoints_mutex); + return PTR_ERR(old); + } + tracepoint_add_old_probes(old); + mutex_unlock(&tracepoints_mutex); + return 0; +} +EXPORT_SYMBOL_GPL(tracepoint_probe_unregister_noupdate); + +/** + * tracepoint_probe_update_all - update tracepoints + */ +void tracepoint_probe_update_all(void) +{ + LIST_HEAD(release_probes); + struct tp_probes *pos, *next; + + mutex_lock(&tracepoints_mutex); + if (!need_update) { + mutex_unlock(&tracepoints_mutex); + return; + } + if (!list_empty(&old_probes)) + list_replace_init(&old_probes, &release_probes); + need_update = 0; + mutex_unlock(&tracepoints_mutex); + + tracepoint_update_probes(); + list_for_each_entry_safe(pos, next, &release_probes, u.list) { + list_del(&pos->u.list); + call_rcu_sched(&pos->u.rcu, rcu_free_old_probes); + } +} +EXPORT_SYMBOL_GPL(tracepoint_probe_update_all); /** * tracepoint_get_iter_range - Get a next tracepoint iterator given a range. @@ -483,3 +541,36 @@ void tracepoint_iter_reset(struct tracepoint_iter *iter) iter->tracepoint = NULL; } EXPORT_SYMBOL_GPL(tracepoint_iter_reset); + +#ifdef CONFIG_MODULES + +int tracepoint_module_notify(struct notifier_block *self, + unsigned long val, void *data) +{ + struct module *mod = data; + + switch (val) { + case MODULE_STATE_COMING: + tracepoint_update_probe_range(mod->tracepoints, + mod->tracepoints + mod->num_tracepoints); + break; + case MODULE_STATE_GOING: + tracepoint_update_probe_range(mod->tracepoints, + mod->tracepoints + mod->num_tracepoints); + break; + } + return 0; +} + +struct notifier_block tracepoint_module_nb = { + .notifier_call = tracepoint_module_notify, + .priority = 0, +}; + +static int init_tracepoints(void) +{ + return register_module_notifier(&tracepoint_module_nb); +} +__initcall(init_tracepoints); + +#endif /* CONFIG_MODULES */ diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index b0f239e443b..1e3fd3e3436 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -545,6 +545,16 @@ config DEBUG_SG If unsure, say N. +config DEBUG_NOTIFIERS + bool "Debug notifier call chains" + depends on DEBUG_KERNEL + help + Enable this to turn on sanity checking for notifier call chains. + This is most useful for kernel developers to make sure that + modules properly unregister themselves from notifier chains. + This is a relatively cheap check but if you care about maximum + performance, say N. + config FRAME_POINTER bool "Compile the kernel with frame pointers" depends on DEBUG_KERNEL && \ diff --git a/mm/memory.c b/mm/memory.c index 164951c4730..fc031d68327 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -3049,3 +3049,18 @@ void print_vma_addr(char *prefix, unsigned long ip) } up_read(¤t->mm->mmap_sem); } + +#ifdef CONFIG_PROVE_LOCKING +void might_fault(void) +{ + might_sleep(); + /* + * it would be nicer only to annotate paths which are not under + * pagefault_disable, however that requires a larger audit and + * providing helpers like get_user_atomic. + */ + if (!in_atomic() && current->mm) + might_lock_read(¤t->mm->mmap_sem); +} +EXPORT_SYMBOL(might_fault); +#endif diff --git a/samples/tracepoints/tp-samples-trace.h b/samples/tracepoints/tp-samples-trace.h index 0216b55bd64..01724e04c55 100644 --- a/samples/tracepoints/tp-samples-trace.h +++ b/samples/tracepoints/tp-samples-trace.h @@ -4,10 +4,10 @@ #include <linux/proc_fs.h> /* for struct inode and struct file */ #include <linux/tracepoint.h> -DEFINE_TRACE(subsys_event, +DECLARE_TRACE(subsys_event, TPPROTO(struct inode *inode, struct file *file), TPARGS(inode, file)); -DEFINE_TRACE(subsys_eventb, +DECLARE_TRACE(subsys_eventb, TPPROTO(void), TPARGS()); #endif diff --git a/samples/tracepoints/tracepoint-probe-sample.c b/samples/tracepoints/tracepoint-probe-sample.c index 55abfdda4bd..e3a964889dc 100644 --- a/samples/tracepoints/tracepoint-probe-sample.c +++ b/samples/tracepoints/tracepoint-probe-sample.c @@ -46,6 +46,7 @@ void __exit tp_sample_trace_exit(void) { unregister_trace_subsys_eventb(probe_subsys_eventb); unregister_trace_subsys_event(probe_subsys_event); + tracepoint_synchronize_unregister(); } module_exit(tp_sample_trace_exit); diff --git a/samples/tracepoints/tracepoint-probe-sample2.c b/samples/tracepoints/tracepoint-probe-sample2.c index 5e9fcf4afff..685a5acb456 100644 --- a/samples/tracepoints/tracepoint-probe-sample2.c +++ b/samples/tracepoints/tracepoint-probe-sample2.c @@ -33,6 +33,7 @@ module_init(tp_sample_trace_init); void __exit tp_sample_trace_exit(void) { unregister_trace_subsys_event(probe_subsys_event); + tracepoint_synchronize_unregister(); } module_exit(tp_sample_trace_exit); diff --git a/samples/tracepoints/tracepoint-sample.c b/samples/tracepoints/tracepoint-sample.c index 4ae4b7fcc04..00d169792a3 100644 --- a/samples/tracepoints/tracepoint-sample.c +++ b/samples/tracepoints/tracepoint-sample.c @@ -13,6 +13,9 @@ #include <linux/proc_fs.h> #include "tp-samples-trace.h" +DEFINE_TRACE(subsys_event); +DEFINE_TRACE(subsys_eventb); + struct proc_dir_entry *pentry_example; static int my_open(struct inode *inode, struct file *file) diff --git a/scripts/Makefile.build b/scripts/Makefile.build index 468fbc9016c..7a176773af8 100644 --- a/scripts/Makefile.build +++ b/scripts/Makefile.build @@ -198,16 +198,10 @@ cmd_modversions = \ fi; endif -ifdef CONFIG_64BIT -arch_bits = 64 -else -arch_bits = 32 -endif - ifdef CONFIG_FTRACE_MCOUNT_RECORD -cmd_record_mcount = perl $(srctree)/scripts/recordmcount.pl \ - "$(ARCH)" "$(arch_bits)" "$(OBJDUMP)" "$(OBJCOPY)" "$(CC)" "$(LD)" \ - "$(NM)" "$(RM)" "$(MV)" "$(@)"; +cmd_record_mcount = perl $(srctree)/scripts/recordmcount.pl "$(ARCH)" \ + "$(if $(CONFIG_64BIT),64,32)" \ + "$(OBJDUMP)" "$(OBJCOPY)" "$(CC)" "$(LD)" "$(NM)" "$(RM)" "$(MV)" "$(@)"; endif define rule_cc_o_c diff --git a/scripts/bootgraph.pl b/scripts/bootgraph.pl index d2c61efc216..f0af9aa9b24 100644 --- a/scripts/bootgraph.pl +++ b/scripts/bootgraph.pl @@ -78,11 +78,13 @@ while (<>) { } if ($count == 0) { - print "No data found in the dmesg. Make sure that 'printk.time=1' and\n"; - print "'initcall_debug' are passed on the kernel command line.\n\n"; - print "Usage: \n"; - print " dmesg | perl scripts/bootgraph.pl > output.svg\n\n"; - exit; + print STDERR <<END; +No data found in the dmesg. Make sure that 'printk.time=1' and +'initcall_debug' are passed on the kernel command line. +Usage: + dmesg | perl scripts/bootgraph.pl > output.svg +END + exit 1; } print "<?xml version=\"1.0\" standalone=\"no\"?> \n"; @@ -109,8 +111,8 @@ my $stylecounter = 0; my %rows; my $rowscount = 1; my @initcalls = sort { $start{$a} <=> $start{$b} } keys(%start); -my $key; -foreach $key (@initcalls) { + +foreach my $key (@initcalls) { my $duration = $end{$key} - $start{$key}; if ($duration >= $threshold) { diff --git a/scripts/recordmcount.pl b/scripts/recordmcount.pl index 6b9fe3eb836..0197e2f6b54 100755 --- a/scripts/recordmcount.pl +++ b/scripts/recordmcount.pl @@ -130,10 +130,13 @@ my %weak; # List of weak functions my %convert; # List of local functions used that needs conversion my $type; +my $nm_regex; # Find the local functions (return function) my $section_regex; # Find the start of a section my $function_regex; # Find the name of a function # (return offset and func name) my $mcount_regex; # Find the call site to mcount (return offset) +my $alignment; # The .align value to use for $mcount_section +my $section_type; # Section header plus possible alignment command if ($arch eq "x86") { if ($bits == 64) { @@ -143,11 +146,21 @@ if ($arch eq "x86") { } } +# +# We base the defaults off of i386, the other archs may +# feel free to change them in the below if statements. +# +$nm_regex = "^[0-9a-fA-F]+\\s+t\\s+(\\S+)"; +$section_regex = "Disassembly of section\\s+(\\S+):"; +$function_regex = "^([0-9a-fA-F]+)\\s+<(.*?)>:"; +$mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\smcount\$"; +$section_type = '@progbits'; +$type = ".long"; + if ($arch eq "x86_64") { - $section_regex = "Disassembly of section\\s+(\\S+):"; - $function_regex = "^([0-9a-fA-F]+)\\s+<(.*?)>:"; $mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\smcount([+-]0x[0-9a-zA-Z]+)?\$"; $type = ".quad"; + $alignment = 8; # force flags for this arch $ld .= " -m elf_x86_64"; @@ -156,10 +169,7 @@ if ($arch eq "x86_64") { $cc .= " -m64"; } elsif ($arch eq "i386") { - $section_regex = "Disassembly of section\\s+(\\S+):"; - $function_regex = "^([0-9a-fA-F]+)\\s+<(.*?)>:"; - $mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\smcount\$"; - $type = ".long"; + $alignment = 4; # force flags for this arch $ld .= " -m elf_i386"; @@ -167,6 +177,27 @@ if ($arch eq "x86_64") { $objcopy .= " -O elf32-i386"; $cc .= " -m32"; +} elsif ($arch eq "sh") { + $alignment = 2; + + # force flags for this arch + $ld .= " -m shlelf_linux"; + $objcopy .= " -O elf32-sh-linux"; + $cc .= " -m32"; + +} elsif ($arch eq "powerpc") { + $nm_regex = "^[0-9a-fA-F]+\\s+t\\s+(\\.?\\S+)"; + $function_regex = "^([0-9a-fA-F]+)\\s+<(\\.?.*?)>:"; + $mcount_regex = "^\\s*([0-9a-fA-F]+):.*\\s\\.?_mcount\$"; + + if ($bits == 64) { + $type = ".quad"; + } + +} elsif ($arch eq "arm") { + $alignment = 2; + $section_type = '%progbits'; + } else { die "Arch $arch is not supported with CONFIG_FTRACE_MCOUNT_RECORD"; } @@ -236,7 +267,7 @@ if (!$found_version) { # open (IN, "$nm $inputfile|") || die "error running $nm"; while (<IN>) { - if (/^[0-9a-fA-F]+\s+t\s+(\S+)/) { + if (/$nm_regex/) { $locals{$1} = 1; } elsif (/^[0-9a-fA-F]+\s+([wW])\s+(\S+)/) { $weak{$2} = $1; @@ -287,7 +318,8 @@ sub update_funcs if (!$opened) { open(FILE, ">$mcount_s") || die "can't create $mcount_s\n"; $opened = 1; - print FILE "\t.section $mcount_section,\"a\",\@progbits\n"; + print FILE "\t.section $mcount_section,\"a\",$section_type\n"; + print FILE "\t.align $alignment\n" if (defined($alignment)); } printf FILE "\t%s %s + %d\n", $type, $ref_func, $offsets[$i] - $offset; } diff --git a/scripts/tracing/draw_functrace.py b/scripts/tracing/draw_functrace.py new file mode 100644 index 00000000000..902f9a99262 --- /dev/null +++ b/scripts/tracing/draw_functrace.py @@ -0,0 +1,130 @@ +#!/usr/bin/python + +""" +Copyright 2008 (c) Frederic Weisbecker <fweisbec@gmail.com> +Licensed under the terms of the GNU GPL License version 2 + +This script parses a trace provided by the function tracer in +kernel/trace/trace_functions.c +The resulted trace is processed into a tree to produce a more human +view of the call stack by drawing textual but hierarchical tree of +calls. Only the functions's names and the the call time are provided. + +Usage: + Be sure that you have CONFIG_FUNCTION_TRACER + # mkdir /debugfs + # mount -t debug debug /debug + # echo function > /debug/tracing/current_tracer + $ cat /debug/tracing/trace_pipe > ~/raw_trace_func + Wait some times but not too much, the script is a bit slow. + Break the pipe (Ctrl + Z) + $ scripts/draw_functrace.py < raw_trace_func > draw_functrace + Then you have your drawn trace in draw_functrace +""" + + +import sys, re + +class CallTree: + """ This class provides a tree representation of the functions + call stack. If a function has no parent in the kernel (interrupt, + syscall, kernel thread...) then it is attached to a virtual parent + called ROOT. + """ + ROOT = None + + def __init__(self, func, time = None, parent = None): + self._func = func + self._time = time + if parent is None: + self._parent = CallTree.ROOT + else: + self._parent = parent + self._children = [] + + def calls(self, func, calltime): + """ If a function calls another one, call this method to insert it + into the tree at the appropriate place. + @return: A reference to the newly created child node. + """ + child = CallTree(func, calltime, self) + self._children.append(child) + return child + + def getParent(self, func): + """ Retrieve the last parent of the current node that + has the name given by func. If this function is not + on a parent, then create it as new child of root + @return: A reference to the parent. + """ + tree = self + while tree != CallTree.ROOT and tree._func != func: + tree = tree._parent + if tree == CallTree.ROOT: + child = CallTree.ROOT.calls(func, None) + return child + return tree + + def __repr__(self): + return self.__toString("", True) + + def __toString(self, branch, lastChild): + if self._time is not None: + s = "%s----%s (%s)\n" % (branch, self._func, self._time) + else: + s = "%s----%s\n" % (branch, self._func) + + i = 0 + if lastChild: + branch = branch[:-1] + " " + while i < len(self._children): + if i != len(self._children) - 1: + s += "%s" % self._children[i].__toString(branch +\ + " |", False) + else: + s += "%s" % self._children[i].__toString(branch +\ + " |", True) + i += 1 + return s + +class BrokenLineException(Exception): + """If the last line is not complete because of the pipe breakage, + we want to stop the processing and ignore this line. + """ + pass + +class CommentLineException(Exception): + """ If the line is a comment (as in the beginning of the trace file), + just ignore it. + """ + pass + + +def parseLine(line): + line = line.strip() + if line.startswith("#"): + raise CommentLineException + m = re.match("[^]]+?\\] +([0-9.]+): (\\w+) <-(\\w+)", line) + if m is None: + raise BrokenLineException + return (m.group(1), m.group(2), m.group(3)) + + +def main(): + CallTree.ROOT = CallTree("Root (Nowhere)", None, None) + tree = CallTree.ROOT + + for line in sys.stdin: + try: + calltime, callee, caller = parseLine(line) + except BrokenLineException: + break + except CommentLineException: + continue + tree = tree.getParent(caller) + tree = tree.calls(callee, calltime) + + print CallTree.ROOT + +if __name__ == "__main__": + main() |