From 97e1c18e8d17bd87e1e383b2e9d9fc740332c8e2 Mon Sep 17 00:00:00 2001 From: Mathieu Desnoyers Date: Fri, 18 Jul 2008 12:16:16 -0400 Subject: tracing: Kernel Tracepoints Implementation of kernel tracepoints. Inspired from the Linux Kernel Markers. Allows complete typing verification by declaring both tracing statement inline functions and probe registration/unregistration static inline functions within the same macro "DEFINE_TRACE". No format string is required. See the tracepoint Documentation and Samples patches for usage examples. Taken from the documentation patch : "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." Addition and removal of tracepoints is synchronized by RCU using the scheduler (and preempt_disable) as guarantees to find a quiescent state (this is really RCU "classic"). The update side uses rcu_barrier_sched() with call_rcu_sched() and the read/execute side uses "preempt_disable()/preempt_enable()". We make sure the previous array containing probes, which has been scheduled for deletion by the rcu callback, is indeed freed before we proceed to the next update. It therefore limits the rate of modification of a single tracepoint to one update per RCU period. The objective here is to permit fast batch add/removal of probes on _different_ tracepoints. Changelog : - Use #name ":" #proto as string to identify the tracepoint in the tracepoint table. This will make sure not type mismatch happens due to connexion of a probe with the wrong type to a tracepoint declared with the same name in a different header. - Add tracepoint_entry_free_old. - Change __TO_TRACE to get rid of the 'i' iterator. Masami Hiramatsu : Tested on x86-64. Performance impact of a tracepoint : same as markers, except that it adds about 70 bytes of instructions in an unlikely branch of each instrumented function (the for loop, the stack setup and the function call). It currently adds a memory read, a test and a conditional branch at the instrumentation site (in the hot path). Immediate values will eventually change this into a load immediate, test and branch, which removes the memory read which will make the i-cache impact smaller (changing the memory read for a load immediate removes 3-4 bytes per site on x86_32 (depending on mov prefixes), or 7-8 bytes on x86_64, it also saves the d-cache hit). About the performance impact of tracepoints (which is comparable to markers), even without immediate values optimizations, tests done by Hideo Aoki on ia64 show no regression. His test case was using hackbench on a kernel where scheduler instrumentation (about 5 events in code scheduler code) was added. Quoting Hideo Aoki about Markers : I evaluated overhead of kernel marker using linux-2.6-sched-fixes git tree, which includes several markers for LTTng, using an ia64 server. While the immediate trace mark feature isn't implemented on ia64, there is no major performance regression. So, I think that we don't have any issues to propose merging marker point patches into Linus's tree from the viewpoint of performance impact. I prepared two kernels to evaluate. The first one was compiled without CONFIG_MARKERS. The second one was enabled CONFIG_MARKERS. I downloaded the original hackbench from the following URL: http://devresources.linux-foundation.org/craiger/hackbench/src/hackbench.c I ran hackbench 5 times in each condition and calculated the average and difference between the kernels. The parameter of hackbench: every 50 from 50 to 800 The number of CPUs of the server: 2, 4, and 8 Below is the results. As you can see, major performance regression wasn't found in any case. Even if number of processes increases, differences between marker-enabled kernel and marker- disabled kernel doesn't increase. Moreover, if number of CPUs increases, the differences doesn't increase either. Curiously, marker-enabled kernel is better than marker-disabled kernel in more than half cases, although I guess it comes from the difference of memory access pattern. * 2 CPUs Number of | without | with | diff | diff | processes | Marker [Sec] | Marker [Sec] | [Sec] | [%] | -------------------------------------------------------------- 50 | 4.811 | 4.872 | +0.061 | +1.27 | 100 | 9.854 | 10.309 | +0.454 | +4.61 | 150 | 15.602 | 15.040 | -0.562 | -3.6 | 200 | 20.489 | 20.380 | -0.109 | -0.53 | 250 | 25.798 | 25.652 | -0.146 | -0.56 | 300 | 31.260 | 30.797 | -0.463 | -1.48 | 350 | 36.121 | 35.770 | -0.351 | -0.97 | 400 | 42.288 | 42.102 | -0.186 | -0.44 | 450 | 47.778 | 47.253 | -0.526 | -1.1 | 500 | 51.953 | 52.278 | +0.325 | +0.63 | 550 | 58.401 | 57.700 | -0.701 | -1.2 | 600 | 63.334 | 63.222 | -0.112 | -0.18 | 650 | 68.816 | 68.511 | -0.306 | -0.44 | 700 | 74.667 | 74.088 | -0.579 | -0.78 | 750 | 78.612 | 79.582 | +0.970 | +1.23 | 800 | 85.431 | 85.263 | -0.168 | -0.2 | -------------------------------------------------------------- * 4 CPUs Number of | without | with | diff | diff | processes | Marker [Sec] | Marker [Sec] | [Sec] | [%] | -------------------------------------------------------------- 50 | 2.586 | 2.584 | -0.003 | -0.1 | 100 | 5.254 | 5.283 | +0.030 | +0.56 | 150 | 8.012 | 8.074 | +0.061 | +0.76 | 200 | 11.172 | 11.000 | -0.172 | -1.54 | 250 | 13.917 | 14.036 | +0.119 | +0.86 | 300 | 16.905 | 16.543 | -0.362 | -2.14 | 350 | 19.901 | 20.036 | +0.135 | +0.68 | 400 | 22.908 | 23.094 | +0.186 | +0.81 | 450 | 26.273 | 26.101 | -0.172 | -0.66 | 500 | 29.554 | 29.092 | -0.461 | -1.56 | 550 | 32.377 | 32.274 | -0.103 | -0.32 | 600 | 35.855 | 35.322 | -0.533 | -1.49 | 650 | 39.192 | 38.388 | -0.804 | -2.05 | 700 | 41.744 | 41.719 | -0.025 | -0.06 | 750 | 45.016 | 44.496 | -0.520 | -1.16 | 800 | 48.212 | 47.603 | -0.609 | -1.26 | -------------------------------------------------------------- * 8 CPUs Number of | without | with | diff | diff | processes | Marker [Sec] | Marker [Sec] | [Sec] | [%] | -------------------------------------------------------------- 50 | 2.094 | 2.072 | -0.022 | -1.07 | 100 | 4.162 | 4.273 | +0.111 | +2.66 | 150 | 6.485 | 6.540 | +0.055 | +0.84 | 200 | 8.556 | 8.478 | -0.078 | -0.91 | 250 | 10.458 | 10.258 | -0.200 | -1.91 | 300 | 12.425 | 12.750 | +0.325 | +2.62 | 350 | 14.807 | 14.839 | +0.032 | +0.22 | 400 | 16.801 | 16.959 | +0.158 | +0.94 | 450 | 19.478 | 19.009 | -0.470 | -2.41 | 500 | 21.296 | 21.504 | +0.208 | +0.98 | 550 | 23.842 | 23.979 | +0.137 | +0.57 | 600 | 26.309 | 26.111 | -0.198 | -0.75 | 650 | 28.705 | 28.446 | -0.259 | -0.9 | 700 | 31.233 | 31.394 | +0.161 | +0.52 | 750 | 34.064 | 33.720 | -0.344 | -1.01 | 800 | 36.320 | 36.114 | -0.206 | -0.57 | -------------------------------------------------------------- Signed-off-by: Mathieu Desnoyers Acked-by: Masami Hiramatsu Acked-by: 'Peter Zijlstra' Signed-off-by: Ingo Molnar --- include/linux/module.h | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) (limited to 'include/linux/module.h') diff --git a/include/linux/module.h b/include/linux/module.h index 68e09557c951..8b6113503863 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -16,6 +16,7 @@ #include #include #include +#include #include #include @@ -331,6 +332,10 @@ struct module struct marker *markers; unsigned int num_markers; #endif +#ifdef CONFIG_TRACEPOINTS + struct tracepoint *tracepoints; + unsigned int num_tracepoints; +#endif #ifdef CONFIG_MODULE_UNLOAD /* What modules depend on me? */ @@ -454,6 +459,9 @@ extern void print_modules(void); extern void module_update_markers(void); +extern void module_update_tracepoints(void); +extern int module_get_iter_tracepoints(struct tracepoint_iter *iter); + #else /* !CONFIG_MODULES... */ #define EXPORT_SYMBOL(sym) #define EXPORT_SYMBOL_GPL(sym) @@ -558,6 +566,15 @@ static inline void module_update_markers(void) { } +static inline void module_update_tracepoints(void) +{ +} + +static inline int module_get_iter_tracepoints(struct tracepoint_iter *iter) +{ + return 0; +} + #endif /* CONFIG_MODULES */ struct device_driver; -- cgit v1.2.3 From 346e15beb5343c2eb8216d820f2ed8f150822b08 Mon Sep 17 00:00:00 2001 From: Jason Baron Date: Tue, 12 Aug 2008 16:46:19 -0400 Subject: driver core: basic infrastructure for per-module dynamic debug messages Base infrastructure to enable per-module debug messages. I've introduced CONFIG_DYNAMIC_PRINTK_DEBUG, which when enabled centralizes control of debugging statements on a per-module basis in one /proc file, currently, /dynamic_printk/modules. When, CONFIG_DYNAMIC_PRINTK_DEBUG, is not set, debugging statements can still be enabled as before, often by defining 'DEBUG' for the proper compilation unit. Thus, this patch set has no affect when CONFIG_DYNAMIC_PRINTK_DEBUG is not set. The infrastructure currently ties into all pr_debug() and dev_dbg() calls. That is, if CONFIG_DYNAMIC_PRINTK_DEBUG is set, all pr_debug() and dev_dbg() calls can be dynamically enabled/disabled on a per-module basis. Future plans include extending this functionality to subsystems, that define their own debug levels and flags. Usage: Dynamic debugging is controlled by the debugfs file, /dynamic_printk/modules. This file contains a list of the modules that can be enabled. The format of the file is as follows: . . . : Name of the module in which the debug call resides : whether the messages are enabled or not For example: snd_hda_intel enabled=0 fixup enabled=1 driver enabled=0 Enable a module: $echo "set enabled=1 " > dynamic_printk/modules Disable a module: $echo "set enabled=0 " > dynamic_printk/modules Enable all modules: $echo "set enabled=1 all" > dynamic_printk/modules Disable all modules: $echo "set enabled=0 all" > dynamic_printk/modules Finally, passing "dynamic_printk" at the command line enables debugging for all modules. This mode can be turned off via the above disable command. [gkh: minor cleanups and tweaks to make the build work quietly] Signed-off-by: Jason Baron Signed-off-by: Greg Kroah-Hartman --- Documentation/kernel-parameters.txt | 5 + include/asm-generic/vmlinux.lds.h | 10 +- include/linux/device.h | 6 +- include/linux/dynamic_printk.h | 93 ++++++++ include/linux/kernel.h | 7 +- include/linux/module.h | 1 - kernel/module.c | 31 +++ lib/Kconfig.debug | 55 +++++ lib/Makefile | 2 + lib/dynamic_printk.c | 418 ++++++++++++++++++++++++++++++++++++ net/netfilter/nf_conntrack_pptp.c | 2 +- scripts/Makefile.lib | 11 +- scripts/basic/Makefile | 2 +- scripts/basic/hash.c | 64 ++++++ 14 files changed, 700 insertions(+), 7 deletions(-) create mode 100644 include/linux/dynamic_printk.h create mode 100644 lib/dynamic_printk.c create mode 100644 scripts/basic/hash.c (limited to 'include/linux/module.h') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 2443f5bb4364..b429c84ceef2 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1713,6 +1713,11 @@ and is between 256 and 4096 characters. It is defined in the file autoconfiguration. Ranges are in pairs (memory base and size). + dynamic_printk + Enables pr_debug()/dev_dbg() calls if + CONFIG_DYNAMIC_PRINTK_DEBUG has been enabled. These can also + be switched on/off via /dynamic_printk/modules + print-fatal-signals= [KNL] debug: print fatal signals print-fatal-signals=1: print segfault info to diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 7440a0dceddb..74c5faf26c05 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -268,7 +268,15 @@ CPU_DISCARD(init.data) \ CPU_DISCARD(init.rodata) \ MEM_DISCARD(init.data) \ - MEM_DISCARD(init.rodata) + MEM_DISCARD(init.rodata) \ + /* implement dynamic printk debug */ \ + VMLINUX_SYMBOL(__start___verbose_strings) = .; \ + *(__verbose_strings) \ + VMLINUX_SYMBOL(__stop___verbose_strings) = .; \ + . = ALIGN(8); \ + VMLINUX_SYMBOL(__start___verbose) = .; \ + *(__verbose) \ + VMLINUX_SYMBOL(__stop___verbose) = .; #define INIT_TEXT \ *(.init.text) \ diff --git a/include/linux/device.h b/include/linux/device.h index 60f6456691a6..fb034461b395 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -550,7 +550,11 @@ extern const char *dev_driver_string(const struct device *dev); #define dev_info(dev, format, arg...) \ dev_printk(KERN_INFO , dev , format , ## arg) -#ifdef DEBUG +#if defined(CONFIG_DYNAMIC_PRINTK_DEBUG) +#define dev_dbg(dev, format, ...) do { \ + dynamic_dev_dbg(dev, format, ##__VA_ARGS__); \ + } while (0) +#elif defined(DEBUG) #define dev_dbg(dev, format, arg...) \ dev_printk(KERN_DEBUG , dev , format , ## arg) #else diff --git a/include/linux/dynamic_printk.h b/include/linux/dynamic_printk.h new file mode 100644 index 000000000000..2d528d009074 --- /dev/null +++ b/include/linux/dynamic_printk.h @@ -0,0 +1,93 @@ +#ifndef _DYNAMIC_PRINTK_H +#define _DYNAMIC_PRINTK_H + +#define DYNAMIC_DEBUG_HASH_BITS 6 +#define DEBUG_HASH_TABLE_SIZE (1 << DYNAMIC_DEBUG_HASH_BITS) + +#define TYPE_BOOLEAN 1 + +#define DYNAMIC_ENABLED_ALL 0 +#define DYNAMIC_ENABLED_NONE 1 +#define DYNAMIC_ENABLED_SOME 2 + +extern int dynamic_enabled; + +/* dynamic_printk_enabled, and dynamic_printk_enabled2 are bitmasks in which + * bit n is set to 1 if any modname hashes into the bucket n, 0 otherwise. They + * use independent hash functions, to reduce the chance of false positives. + */ +extern long long dynamic_printk_enabled; +extern long long dynamic_printk_enabled2; + +struct mod_debug { + char *modname; + char *logical_modname; + char *flag_names; + int type; + int hash; + int hash2; +} __attribute__((aligned(8))); + +int register_dynamic_debug_module(char *mod_name, int type, char *share_name, + char *flags, int hash, int hash2); + +#if defined(CONFIG_DYNAMIC_PRINTK_DEBUG) +extern int unregister_dynamic_debug_module(char *mod_name); +extern int __dynamic_dbg_enabled_helper(char *modname, int type, + int value, int hash); + +#define __dynamic_dbg_enabled(module, type, value, level, hash) ({ \ + int __ret = 0; \ + if (unlikely((dynamic_printk_enabled & (1LL << DEBUG_HASH)) && \ + (dynamic_printk_enabled2 & (1LL << DEBUG_HASH2)))) \ + __ret = __dynamic_dbg_enabled_helper(module, type, \ + value, hash);\ + __ret; }) + +#define dynamic_pr_debug(fmt, ...) do { \ + static char mod_name[] \ + __attribute__((section("__verbose_strings"))) \ + = KBUILD_MODNAME; \ + static struct mod_debug descriptor \ + __used \ + __attribute__((section("__verbose"), aligned(8))) = \ + { mod_name, mod_name, NULL, TYPE_BOOLEAN, DEBUG_HASH, DEBUG_HASH2 };\ + if (__dynamic_dbg_enabled(KBUILD_MODNAME, TYPE_BOOLEAN, \ + 0, 0, DEBUG_HASH)) \ + printk(KERN_DEBUG KBUILD_MODNAME ":" fmt, \ + ##__VA_ARGS__); \ + } while (0) + +#define dynamic_dev_dbg(dev, format, ...) do { \ + static char mod_name[] \ + __attribute__((section("__verbose_strings"))) \ + = KBUILD_MODNAME; \ + static struct mod_debug descriptor \ + __used \ + __attribute__((section("__verbose"), aligned(8))) = \ + { mod_name, mod_name, NULL, TYPE_BOOLEAN, DEBUG_HASH, DEBUG_HASH2 };\ + if (__dynamic_dbg_enabled(KBUILD_MODNAME, TYPE_BOOLEAN, \ + 0, 0, DEBUG_HASH)) \ + dev_printk(KERN_DEBUG, dev, \ + KBUILD_MODNAME ": " format, \ + ##__VA_ARGS__); \ + } while (0) + +#else + +static inline int unregister_dynamic_debug_module(const char *mod_name) +{ + return 0; +} +static inline int __dynamic_dbg_enabled_helper(char *modname, int type, + int value, int hash) +{ + return 0; +} + +#define __dynamic_dbg_enabled(module, type, value, level, hash) ({ 0; }) +#define dynamic_pr_debug(fmt, ...) do { } while (0) +#define dynamic_dev_dbg(dev, format, ...) do { } while (0) +#endif + +#endif diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 75d81f157d2e..ededb6e83b41 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -16,6 +16,7 @@ #include #include #include +#include #include #include @@ -303,8 +304,12 @@ static inline char *pack_hex_byte(char *buf, u8 byte) #define pr_info(fmt, arg...) \ printk(KERN_INFO fmt, ##arg) -#ifdef DEBUG /* If you are writing a driver, please use dev_dbg instead */ +#if defined(CONFIG_DYNAMIC_PRINTK_DEBUG) +#define pr_debug(fmt, ...) do { \ + dynamic_pr_debug(fmt, ##__VA_ARGS__); \ + } while (0) +#elif defined(DEBUG) #define pr_debug(fmt, arg...) \ printk(KERN_DEBUG fmt, ##arg) #else diff --git a/include/linux/module.h b/include/linux/module.h index 68e09557c951..a41555cbe00a 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -345,7 +345,6 @@ struct module /* Reference counts */ struct module_ref ref[NR_CPUS]; #endif - }; #ifndef MODULE_ARCH_INIT #define MODULE_ARCH_INIT {} diff --git a/kernel/module.c b/kernel/module.c index d5fcd24e5aff..c52700667292 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -784,6 +784,7 @@ sys_delete_module(const char __user *name_user, unsigned int flags) mutex_lock(&module_mutex); /* Store the name of the last unloaded module for diagnostic purposes */ strlcpy(last_unloaded_module, mod->name, sizeof(last_unloaded_module)); + unregister_dynamic_debug_module(mod->name); free_module(mod); out: @@ -1783,6 +1784,33 @@ static inline void add_kallsyms(struct module *mod, } #endif /* CONFIG_KALLSYMS */ +#ifdef CONFIG_DYNAMIC_PRINTK_DEBUG +static void dynamic_printk_setup(Elf_Shdr *sechdrs, unsigned int verboseindex) +{ + struct mod_debug *debug_info; + unsigned long pos, end; + unsigned int num_verbose; + + pos = sechdrs[verboseindex].sh_addr; + num_verbose = sechdrs[verboseindex].sh_size / + sizeof(struct mod_debug); + end = pos + (num_verbose * sizeof(struct mod_debug)); + + for (; pos < end; pos += sizeof(struct mod_debug)) { + debug_info = (struct mod_debug *)pos; + register_dynamic_debug_module(debug_info->modname, + debug_info->type, debug_info->logical_modname, + debug_info->flag_names, debug_info->hash, + debug_info->hash2); + } +} +#else +static inline void dynamic_printk_setup(Elf_Shdr *sechdrs, + unsigned int verboseindex) +{ +} +#endif /* CONFIG_DYNAMIC_PRINTK_DEBUG */ + static void *module_alloc_update_bounds(unsigned long size) { void *ret = module_alloc(size); @@ -1831,6 +1859,7 @@ static noinline struct module *load_module(void __user *umod, #endif unsigned int markersindex; unsigned int markersstringsindex; + unsigned int verboseindex; struct module *mod; long err = 0; void *percpu = NULL, *ptr = NULL; /* Stops spurious gcc warning */ @@ -2117,6 +2146,7 @@ static noinline struct module *load_module(void __user *umod, markersindex = find_sec(hdr, sechdrs, secstrings, "__markers"); markersstringsindex = find_sec(hdr, sechdrs, secstrings, "__markers_strings"); + verboseindex = find_sec(hdr, sechdrs, secstrings, "__verbose"); /* Now do relocations. */ for (i = 1; i < hdr->e_shnum; i++) { @@ -2167,6 +2197,7 @@ static noinline struct module *load_module(void __user *umod, marker_update_probe_range(mod->markers, mod->markers + mod->num_markers); #endif + dynamic_printk_setup(sechdrs, verboseindex); err = module_finalize(hdr, sechdrs, mod); if (err < 0) goto cleanup; diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index aa81d2848448..31d784dd80d0 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -807,6 +807,61 @@ menuconfig BUILD_DOCSRC Say N if you are unsure. +config DYNAMIC_PRINTK_DEBUG + bool "Enable dynamic printk() call support" + default n + depends on PRINTK + select PRINTK_DEBUG + help + + Compiles debug level messages into the kernel, which would not + otherwise be available at runtime. These messages can then be + enabled/disabled on a per module basis. This mechanism implicitly + enables all pr_debug() and dev_dbg() calls. The impact of this + compile option is a larger kernel text size of about 2%. + + Usage: + + Dynamic debugging is controlled by the debugfs file, + dynamic_printk/modules. This file contains a list of the modules that + can be enabled. The format of the file is the module name, followed + by a set of flags that can be enabled. The first flag is always the + 'enabled' flag. For example: + + + . + . + . + + : Name of the module in which the debug call resides + : whether the messages are enabled or not + + From a live system: + + snd_hda_intel enabled=0 + fixup enabled=0 + driver enabled=0 + + Enable a module: + + $echo "set enabled=1 " > dynamic_printk/modules + + Disable a module: + + $echo "set enabled=0 " > dynamic_printk/modules + + Enable all modules: + + $echo "set enabled=1 all" > dynamic_printk/modules + + Disable all modules: + + $echo "set enabled=0 all" > dynamic_printk/modules + + Finally, passing "dynamic_printk" at the command line enables + debugging for all modules. This mode can be turned off via the above + disable command. + source "samples/Kconfig" source "lib/Kconfig.kgdb" diff --git a/lib/Makefile b/lib/Makefile index 44001af76a7d..16feaab057b2 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -81,6 +81,8 @@ obj-$(CONFIG_HAVE_LMB) += lmb.o obj-$(CONFIG_HAVE_ARCH_TRACEHOOK) += syscall.o +obj-$(CONFIG_DYNAMIC_PRINTK_DEBUG) += dynamic_printk.o + hostprogs-y := gen_crc32table clean-files := crc32table.h diff --git a/lib/dynamic_printk.c b/lib/dynamic_printk.c new file mode 100644 index 000000000000..d640f87bdc9e --- /dev/null +++ b/lib/dynamic_printk.c @@ -0,0 +1,418 @@ +/* + * lib/dynamic_printk.c + * + * make pr_debug()/dev_dbg() calls runtime configurable based upon their + * their source module. + * + * Copyright (C) 2008 Red Hat, Inc., Jason Baron + */ + +#include +#include +#include +#include +#include +#include + +extern struct mod_debug __start___verbose[]; +extern struct mod_debug __stop___verbose[]; + +struct debug_name { + struct hlist_node hlist; + struct hlist_node hlist2; + int hash1; + int hash2; + char *name; + int enable; + int type; +}; + +static int nr_entries; +static int num_enabled; +int dynamic_enabled = DYNAMIC_ENABLED_NONE; +static struct hlist_head module_table[DEBUG_HASH_TABLE_SIZE] = + { [0 ... DEBUG_HASH_TABLE_SIZE-1] = HLIST_HEAD_INIT }; +static struct hlist_head module_table2[DEBUG_HASH_TABLE_SIZE] = + { [0 ... DEBUG_HASH_TABLE_SIZE-1] = HLIST_HEAD_INIT }; +static DECLARE_MUTEX(debug_list_mutex); + +/* dynamic_printk_enabled, and dynamic_printk_enabled2 are bitmasks in which + * bit n is set to 1 if any modname hashes into the bucket n, 0 otherwise. They + * use independent hash functions, to reduce the chance of false positives. + */ +long long dynamic_printk_enabled; +EXPORT_SYMBOL_GPL(dynamic_printk_enabled); +long long dynamic_printk_enabled2; +EXPORT_SYMBOL_GPL(dynamic_printk_enabled2); + +/* returns the debug module pointer. */ +static struct debug_name *find_debug_module(char *module_name) +{ + int i; + struct hlist_head *head; + struct hlist_node *node; + struct debug_name *element; + + element = NULL; + for (i = 0; i < DEBUG_HASH_TABLE_SIZE; i++) { + head = &module_table[i]; + hlist_for_each_entry_rcu(element, node, head, hlist) + if (!strcmp(element->name, module_name)) + return element; + } + return NULL; +} + +/* returns the debug module pointer. */ +static struct debug_name *find_debug_module_hash(char *module_name, int hash) +{ + struct hlist_head *head; + struct hlist_node *node; + struct debug_name *element; + + element = NULL; + head = &module_table[hash]; + hlist_for_each_entry_rcu(element, node, head, hlist) + if (!strcmp(element->name, module_name)) + return element; + return NULL; +} + +/* caller must hold mutex*/ +static int __add_debug_module(char *mod_name, int hash, int hash2) +{ + struct debug_name *new; + char *module_name; + int ret = 0; + + if (find_debug_module(mod_name)) { + ret = -EINVAL; + goto out; + } + module_name = kmalloc(strlen(mod_name) + 1, GFP_KERNEL); + if (!module_name) { + ret = -ENOMEM; + goto out; + } + module_name = strcpy(module_name, mod_name); + module_name[strlen(mod_name)] = '\0'; + new = kzalloc(sizeof(struct debug_name), GFP_KERNEL); + if (!new) { + kfree(module_name); + ret = -ENOMEM; + goto out; + } + INIT_HLIST_NODE(&new->hlist); + INIT_HLIST_NODE(&new->hlist2); + new->name = module_name; + new->hash1 = hash; + new->hash2 = hash2; + hlist_add_head_rcu(&new->hlist, &module_table[hash]); + hlist_add_head_rcu(&new->hlist2, &module_table2[hash2]); + nr_entries++; +out: + return ret; +} + +int unregister_dynamic_debug_module(char *mod_name) +{ + struct debug_name *element; + int ret = 0; + + down(&debug_list_mutex); + element = find_debug_module(mod_name); + if (!element) { + ret = -EINVAL; + goto out; + } + hlist_del_rcu(&element->hlist); + hlist_del_rcu(&element->hlist2); + synchronize_rcu(); + kfree(element->name); + if (element->enable) + num_enabled--; + kfree(element); + nr_entries--; +out: + up(&debug_list_mutex); + return 0; +} +EXPORT_SYMBOL_GPL(unregister_dynamic_debug_module); + +int register_dynamic_debug_module(char *mod_name, int type, char *share_name, + char *flags, int hash, int hash2) +{ + struct debug_name *elem; + int ret = 0; + + down(&debug_list_mutex); + elem = find_debug_module(mod_name); + if (!elem) { + if (__add_debug_module(mod_name, hash, hash2)) + goto out; + elem = find_debug_module(mod_name); + if (dynamic_enabled == DYNAMIC_ENABLED_ALL && + !strcmp(mod_name, share_name)) { + elem->enable = true; + num_enabled++; + } + } + elem->type |= type; +out: + up(&debug_list_mutex); + return ret; +} +EXPORT_SYMBOL_GPL(register_dynamic_debug_module); + +int __dynamic_dbg_enabled_helper(char *mod_name, int type, int value, int hash) +{ + struct debug_name *elem; + int ret = 0; + + if (dynamic_enabled == DYNAMIC_ENABLED_ALL) + return 1; + rcu_read_lock(); + elem = find_debug_module_hash(mod_name, hash); + if (elem && elem->enable) + ret = 1; + rcu_read_unlock(); + return ret; +} +EXPORT_SYMBOL_GPL(__dynamic_dbg_enabled_helper); + +static void set_all(bool enable) +{ + struct debug_name *e; + struct hlist_node *node; + int i; + long long enable_mask; + + for (i = 0; i < DEBUG_HASH_TABLE_SIZE; i++) { + if (module_table[i].first != NULL) { + hlist_for_each_entry(e, node, &module_table[i], hlist) { + e->enable = enable; + } + } + } + if (enable) + enable_mask = ULLONG_MAX; + else + enable_mask = 0; + dynamic_printk_enabled = enable_mask; + dynamic_printk_enabled2 = enable_mask; +} + +static int disabled_hash(int i, bool first_table) +{ + struct debug_name *e; + struct hlist_node *node; + + if (first_table) { + hlist_for_each_entry(e, node, &module_table[i], hlist) { + if (e->enable) + return 0; + } + } else { + hlist_for_each_entry(e, node, &module_table2[i], hlist2) { + if (e->enable) + return 0; + } + } + return 1; +} + +static ssize_t pr_debug_write(struct file *file, const char __user *buf, + size_t length, loff_t *ppos) +{ + char *buffer, *s, *value_str, *setting_str; + int err, value; + struct debug_name *elem = NULL; + int all = 0; + + if (length > PAGE_SIZE || length < 0) + return -EINVAL; + + buffer = (char *)__get_free_page(GFP_KERNEL); + if (!buffer) + return -ENOMEM; + + err = -EFAULT; + if (copy_from_user(buffer, buf, length)) + goto out; + + err = -EINVAL; + if (length < PAGE_SIZE) + buffer[length] = '\0'; + else if (buffer[PAGE_SIZE-1]) + goto out; + + err = -EINVAL; + down(&debug_list_mutex); + + if (strncmp("set", buffer, 3)) + goto out_up; + s = buffer + 3; + setting_str = strsep(&s, "="); + if (s == NULL) + goto out_up; + setting_str = strstrip(setting_str); + value_str = strsep(&s, " "); + if (s == NULL) + goto out_up; + s = strstrip(s); + if (!strncmp(s, "all", 3)) + all = 1; + else + elem = find_debug_module(s); + if (!strncmp(setting_str, "enable", 6)) { + value = !!simple_strtol(value_str, NULL, 10); + if (all) { + if (value) { + set_all(true); + num_enabled = nr_entries; + dynamic_enabled = DYNAMIC_ENABLED_ALL; + } else { + set_all(false); + num_enabled = 0; + dynamic_enabled = DYNAMIC_ENABLED_NONE; + } + err = 0; + } else { + if (elem) { + if (value && (elem->enable == 0)) { + dynamic_printk_enabled |= + (1LL << elem->hash1); + dynamic_printk_enabled2 |= + (1LL << elem->hash2); + elem->enable = 1; + num_enabled++; + dynamic_enabled = DYNAMIC_ENABLED_SOME; + err = 0; + printk(KERN_DEBUG + "debugging enabled for module %s", + elem->name); + } else if (!value && (elem->enable == 1)) { + elem->enable = 0; + num_enabled--; + if (disabled_hash(elem->hash1, true)) + dynamic_printk_enabled &= + ~(1LL << elem->hash1); + if (disabled_hash(elem->hash2, false)) + dynamic_printk_enabled2 &= + ~(1LL << elem->hash2); + if (num_enabled) + dynamic_enabled = + DYNAMIC_ENABLED_SOME; + else + dynamic_enabled = + DYNAMIC_ENABLED_NONE; + err = 0; + printk(KERN_DEBUG + "debugging disabled for module " + "%s", elem->name); + } + } + } + } + if (!err) + err = length; +out_up: + up(&debug_list_mutex); +out: + free_page((unsigned long)buffer); + return err; +} + +static void *pr_debug_seq_start(struct seq_file *f, loff_t *pos) +{ + return (*pos < DEBUG_HASH_TABLE_SIZE) ? pos : NULL; +} + +static void *pr_debug_seq_next(struct seq_file *s, void *v, loff_t *pos) +{ + (*pos)++; + if (*pos >= DEBUG_HASH_TABLE_SIZE) + return NULL; + return pos; +} + +static void pr_debug_seq_stop(struct seq_file *s, void *v) +{ + /* Nothing to do */ +} + +static int pr_debug_seq_show(struct seq_file *s, void *v) +{ + struct hlist_head *head; + struct hlist_node *node; + struct debug_name *elem; + unsigned int i = *(loff_t *) v; + + rcu_read_lock(); + head = &module_table[i]; + hlist_for_each_entry_rcu(elem, node, head, hlist) { + seq_printf(s, "%s enabled=%d", elem->name, elem->enable); + seq_printf(s, "\n"); + } + rcu_read_unlock(); + return 0; +} + +static struct seq_operations pr_debug_seq_ops = { + .start = pr_debug_seq_start, + .next = pr_debug_seq_next, + .stop = pr_debug_seq_stop, + .show = pr_debug_seq_show +}; + +static int pr_debug_open(struct inode *inode, struct file *filp) +{ + return seq_open(filp, &pr_debug_seq_ops); +} + +static const struct file_operations pr_debug_operations = { + .open = pr_debug_open, + .read = seq_read, + .write = pr_debug_write, + .llseek = seq_lseek, + .release = seq_release, +}; + +static int __init dynamic_printk_init(void) +{ + struct dentry *dir, *file; + struct mod_debug *iter; + unsigned long value; + + dir = debugfs_create_dir("dynamic_printk", NULL); + if (!dir) + return -ENOMEM; + file = debugfs_create_file("modules", 0644, dir, NULL, + &pr_debug_operations); + if (!file) { + debugfs_remove(dir); + return -ENOMEM; + } + for (value = (unsigned long)__start___verbose; + value < (unsigned long)__stop___verbose; + value += sizeof(struct mod_debug)) { + iter = (struct mod_debug *)value; + register_dynamic_debug_module(iter->modname, + iter->type, + iter->logical_modname, + iter->flag_names, iter->hash, iter->hash2); + } + return 0; +} +module_init(dynamic_printk_init); +/* may want to move this earlier so we can get traces as early as possible */ + +static int __init dynamic_printk_setup(char *str) +{ + if (str) + return -ENOENT; + set_all(true); + return 0; +} +/* Use early_param(), so we can get debug output as early as possible */ +early_param("dynamic_printk", dynamic_printk_setup); diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c index 373e51e91ce5..1bc3001d1827 100644 --- a/net/netfilter/nf_conntrack_pptp.c +++ b/net/netfilter/nf_conntrack_pptp.c @@ -65,7 +65,7 @@ void struct nf_conntrack_expect *exp) __read_mostly; EXPORT_SYMBOL_GPL(nf_nat_pptp_hook_expectfn); -#ifdef DEBUG +#if defined(DEBUG) || defined(CONFIG_DYNAMIC_PRINTK_DEBUG) /* PptpControlMessageType names */ const char *const pptp_msg_name[] = { "UNKNOWN_MESSAGE", diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib index ea48b82a3707..b4ca38a21158 100644 --- a/scripts/Makefile.lib +++ b/scripts/Makefile.lib @@ -96,6 +96,14 @@ basename_flags = -D"KBUILD_BASENAME=KBUILD_STR($(call name-fix,$(basetarget)))" modname_flags = $(if $(filter 1,$(words $(modname))),\ -D"KBUILD_MODNAME=KBUILD_STR($(call name-fix,$(modname)))") +#hash values +ifdef CONFIG_DYNAMIC_PRINTK_DEBUG +debug_flags = -D"DEBUG_HASH=$(shell ./scripts/basic/hash djb2 $(@D)$(modname))"\ + -D"DEBUG_HASH2=$(shell ./scripts/basic/hash r5 $(@D)$(modname))" +else +debug_flags = +endif + orig_c_flags = $(KBUILD_CFLAGS) $(ccflags-y) $(CFLAGS_$(basetarget).o) _c_flags = $(filter-out $(CFLAGS_REMOVE_$(basetarget).o), $(orig_c_flags)) _a_flags = $(KBUILD_AFLAGS) $(asflags-y) $(AFLAGS_$(basetarget).o) @@ -121,7 +129,8 @@ endif c_flags = -Wp,-MD,$(depfile) $(NOSTDINC_FLAGS) $(KBUILD_CPPFLAGS) \ $(__c_flags) $(modkern_cflags) \ - -D"KBUILD_STR(s)=\#s" $(basename_flags) $(modname_flags) + -D"KBUILD_STR(s)=\#s" $(basename_flags) $(modname_flags) \ + $(debug_flags) a_flags = -Wp,-MD,$(depfile) $(NOSTDINC_FLAGS) $(KBUILD_CPPFLAGS) \ $(__a_flags) $(modkern_aflags) diff --git a/scripts/basic/Makefile b/scripts/basic/Makefile index 4c324a1f1e0e..09559951df12 100644 --- a/scripts/basic/Makefile +++ b/scripts/basic/Makefile @@ -9,7 +9,7 @@ # fixdep: Used to generate dependency information during build process # docproc: Used in Documentation/DocBook -hostprogs-y := fixdep docproc +hostprogs-y := fixdep docproc hash always := $(hostprogs-y) # fixdep is needed to compile other host programs diff --git a/scripts/basic/hash.c b/scripts/basic/hash.c new file mode 100644 index 000000000000..3299ad7fc8c0 --- /dev/null +++ b/scripts/basic/hash.c @@ -0,0 +1,64 @@ +/* + * Copyright (C) 2008 Red Hat, Inc., Jason Baron + * + */ + +#include +#include +#include + +#define DYNAMIC_DEBUG_HASH_BITS 6 + +static const char *program; + +static void usage(void) +{ + printf("Usage: %s \n", program); + exit(1); +} + +/* djb2 hashing algorithm by Dan Bernstein. From: + * http://www.cse.yorku.ca/~oz/hash.html + */ + +unsigned int djb2_hash(char *str) +{ + unsigned long hash = 5381; + int c; + + c = *str; + while (c) { + hash = ((hash << 5) + hash) + c; + c = *++str; + } + return (unsigned int)(hash & ((1 << DYNAMIC_DEBUG_HASH_BITS) - 1)); +} + +unsigned int r5_hash(char *str) +{ + unsigned long hash = 0; + int c; + + c = *str; + while (c) { + hash = (hash + (c << 4) + (c >> 4)) * 11; + c = *++str; + } + return (unsigned int)(hash & ((1 << DYNAMIC_DEBUG_HASH_BITS) - 1)); +} + +int main(int argc, char *argv[]) +{ + program = argv[0]; + + if (argc != 3) + usage(); + if (!strcmp(argv[1], "djb2")) + printf("%d\n", djb2_hash(argv[2])); + else if (!strcmp(argv[1], "r5")) + printf("%d\n", r5_hash(argv[2])); + else + usage(); + exit(0); +} + -- cgit v1.2.3