From 9183c3f9ed710a8edf1a61e8a96d497258d26e08 Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Tue, 18 Aug 2020 15:57:42 +0200 Subject: static_call: Add inline static call infrastructure Add infrastructure for an arch-specific CONFIG_HAVE_STATIC_CALL_INLINE option, which is a faster version of CONFIG_HAVE_STATIC_CALL. At runtime, the static call sites are patched directly, rather than using the out-of-line trampolines. Compared to out-of-line static calls, the performance benefits are more modest, but still measurable. Steven Rostedt did some tracepoint measurements: https://lkml.kernel.org/r/20181126155405.72b4f718@gandalf.local.home This code is heavily inspired by the jump label code (aka "static jumps"), as some of the concepts are very similar. For more details, see the comments in include/linux/static_call.h. [peterz: simplified interface; merged trampolines] Signed-off-by: Josh Poimboeuf Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Reviewed-by: Steven Rostedt (VMware) Cc: Linus Torvalds Link: https://lore.kernel.org/r/20200818135804.684334440@infradead.org --- kernel/static_call.c | 303 +++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 303 insertions(+) create mode 100644 kernel/static_call.c (limited to 'kernel/static_call.c') diff --git a/kernel/static_call.c b/kernel/static_call.c new file mode 100644 index 000000000000..d24349244675 --- /dev/null +++ b/kernel/static_call.c @@ -0,0 +1,303 @@ +// SPDX-License-Identifier: GPL-2.0 +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +extern struct static_call_site __start_static_call_sites[], + __stop_static_call_sites[]; + +static bool static_call_initialized; + +#define STATIC_CALL_INIT 1UL + +/* mutex to protect key modules/sites */ +static DEFINE_MUTEX(static_call_mutex); + +static void static_call_lock(void) +{ + mutex_lock(&static_call_mutex); +} + +static void static_call_unlock(void) +{ + mutex_unlock(&static_call_mutex); +} + +static inline void *static_call_addr(struct static_call_site *site) +{ + return (void *)((long)site->addr + (long)&site->addr); +} + + +static inline struct static_call_key *static_call_key(const struct static_call_site *site) +{ + return (struct static_call_key *) + (((long)site->key + (long)&site->key) & ~STATIC_CALL_INIT); +} + +/* These assume the key is word-aligned. */ +static inline bool static_call_is_init(struct static_call_site *site) +{ + return ((long)site->key + (long)&site->key) & STATIC_CALL_INIT; +} + +static inline void static_call_set_init(struct static_call_site *site) +{ + site->key = ((long)static_call_key(site) | STATIC_CALL_INIT) - + (long)&site->key; +} + +static int static_call_site_cmp(const void *_a, const void *_b) +{ + const struct static_call_site *a = _a; + const struct static_call_site *b = _b; + const struct static_call_key *key_a = static_call_key(a); + const struct static_call_key *key_b = static_call_key(b); + + if (key_a < key_b) + return -1; + + if (key_a > key_b) + return 1; + + return 0; +} + +static void static_call_site_swap(void *_a, void *_b, int size) +{ + long delta = (unsigned long)_a - (unsigned long)_b; + struct static_call_site *a = _a; + struct static_call_site *b = _b; + struct static_call_site tmp = *a; + + a->addr = b->addr - delta; + a->key = b->key - delta; + + b->addr = tmp.addr + delta; + b->key = tmp.key + delta; +} + +static inline void static_call_sort_entries(struct static_call_site *start, + struct static_call_site *stop) +{ + sort(start, stop - start, sizeof(struct static_call_site), + static_call_site_cmp, static_call_site_swap); +} + +void __static_call_update(struct static_call_key *key, void *tramp, void *func) +{ + struct static_call_site *site, *stop; + struct static_call_mod *site_mod; + + cpus_read_lock(); + static_call_lock(); + + if (key->func == func) + goto done; + + key->func = func; + + arch_static_call_transform(NULL, tramp, func); + + /* + * If uninitialized, we'll not update the callsites, but they still + * point to the trampoline and we just patched that. + */ + if (WARN_ON_ONCE(!static_call_initialized)) + goto done; + + for (site_mod = key->mods; site_mod; site_mod = site_mod->next) { + struct module *mod = site_mod->mod; + + if (!site_mod->sites) { + /* + * This can happen if the static call key is defined in + * a module which doesn't use it. + */ + continue; + } + + stop = __stop_static_call_sites; + +#ifdef CONFIG_MODULES + if (mod) { + stop = mod->static_call_sites + + mod->num_static_call_sites; + } +#endif + + for (site = site_mod->sites; + site < stop && static_call_key(site) == key; site++) { + void *site_addr = static_call_addr(site); + + if (static_call_is_init(site)) { + /* + * Don't write to call sites which were in + * initmem and have since been freed. + */ + if (!mod && system_state >= SYSTEM_RUNNING) + continue; + if (mod && !within_module_init((unsigned long)site_addr, mod)) + continue; + } + + if (!kernel_text_address((unsigned long)site_addr)) { + WARN_ONCE(1, "can't patch static call site at %pS", + site_addr); + continue; + } + + arch_static_call_transform(site_addr, NULL, func); + } + } + +done: + static_call_unlock(); + cpus_read_unlock(); +} +EXPORT_SYMBOL_GPL(__static_call_update); + +static int __static_call_init(struct module *mod, + struct static_call_site *start, + struct static_call_site *stop) +{ + struct static_call_site *site; + struct static_call_key *key, *prev_key = NULL; + struct static_call_mod *site_mod; + + if (start == stop) + return 0; + + static_call_sort_entries(start, stop); + + for (site = start; site < stop; site++) { + void *site_addr = static_call_addr(site); + + if ((mod && within_module_init((unsigned long)site_addr, mod)) || + (!mod && init_section_contains(site_addr, 1))) + static_call_set_init(site); + + key = static_call_key(site); + if (key != prev_key) { + prev_key = key; + + site_mod = kzalloc(sizeof(*site_mod), GFP_KERNEL); + if (!site_mod) + return -ENOMEM; + + site_mod->mod = mod; + site_mod->sites = site; + site_mod->next = key->mods; + key->mods = site_mod; + } + + arch_static_call_transform(site_addr, NULL, key->func); + } + + return 0; +} + +#ifdef CONFIG_MODULES + +static int static_call_add_module(struct module *mod) +{ + return __static_call_init(mod, mod->static_call_sites, + mod->static_call_sites + mod->num_static_call_sites); +} + +static void static_call_del_module(struct module *mod) +{ + struct static_call_site *start = mod->static_call_sites; + struct static_call_site *stop = mod->static_call_sites + + mod->num_static_call_sites; + struct static_call_key *key, *prev_key = NULL; + struct static_call_mod *site_mod, **prev; + struct static_call_site *site; + + for (site = start; site < stop; site++) { + key = static_call_key(site); + if (key == prev_key) + continue; + + prev_key = key; + + for (prev = &key->mods, site_mod = key->mods; + site_mod && site_mod->mod != mod; + prev = &site_mod->next, site_mod = site_mod->next) + ; + + if (!site_mod) + continue; + + *prev = site_mod->next; + kfree(site_mod); + } +} + +static int static_call_module_notify(struct notifier_block *nb, + unsigned long val, void *data) +{ + struct module *mod = data; + int ret = 0; + + cpus_read_lock(); + static_call_lock(); + + switch (val) { + case MODULE_STATE_COMING: + ret = static_call_add_module(mod); + if (ret) { + WARN(1, "Failed to allocate memory for static calls"); + static_call_del_module(mod); + } + break; + case MODULE_STATE_GOING: + static_call_del_module(mod); + break; + } + + static_call_unlock(); + cpus_read_unlock(); + + return notifier_from_errno(ret); +} + +static struct notifier_block static_call_module_nb = { + .notifier_call = static_call_module_notify, +}; + +#endif /* CONFIG_MODULES */ + +static void __init static_call_init(void) +{ + int ret; + + if (static_call_initialized) + return; + + cpus_read_lock(); + static_call_lock(); + ret = __static_call_init(NULL, __start_static_call_sites, + __stop_static_call_sites); + static_call_unlock(); + cpus_read_unlock(); + + if (ret) { + pr_err("Failed to allocate memory for static_call!\n"); + BUG(); + } + + static_call_initialized = true; + +#ifdef CONFIG_MODULES + register_module_notifier(&static_call_module_nb); +#endif +} +early_initcall(static_call_init); -- cgit v1.2.3 From 6333e8f73b834f54e395a056e6002403f0862c51 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 18 Aug 2020 15:57:43 +0200 Subject: static_call: Avoid kprobes on inline static_call()s Similar to how we disallow kprobes on any other dynamic text (ftrace/jump_label) also disallow kprobes on inline static_call()s. Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20200818135804.744920586@infradead.org --- kernel/static_call.c | 68 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 68 insertions(+) (limited to 'kernel/static_call.c') diff --git a/kernel/static_call.c b/kernel/static_call.c index d24349244675..753b2f1b4fb8 100644 --- a/kernel/static_call.c +++ b/kernel/static_call.c @@ -204,8 +204,58 @@ static int __static_call_init(struct module *mod, return 0; } +static int addr_conflict(struct static_call_site *site, void *start, void *end) +{ + unsigned long addr = (unsigned long)static_call_addr(site); + + if (addr <= (unsigned long)end && + addr + CALL_INSN_SIZE > (unsigned long)start) + return 1; + + return 0; +} + +static int __static_call_text_reserved(struct static_call_site *iter_start, + struct static_call_site *iter_stop, + void *start, void *end) +{ + struct static_call_site *iter = iter_start; + + while (iter < iter_stop) { + if (addr_conflict(iter, start, end)) + return 1; + iter++; + } + + return 0; +} + #ifdef CONFIG_MODULES +static int __static_call_mod_text_reserved(void *start, void *end) +{ + struct module *mod; + int ret; + + preempt_disable(); + mod = __module_text_address((unsigned long)start); + WARN_ON_ONCE(__module_text_address((unsigned long)end) != mod); + if (!try_module_get(mod)) + mod = NULL; + preempt_enable(); + + if (!mod) + return 0; + + ret = __static_call_text_reserved(mod->static_call_sites, + mod->static_call_sites + mod->num_static_call_sites, + start, end); + + module_put(mod); + + return ret; +} + static int static_call_add_module(struct module *mod) { return __static_call_init(mod, mod->static_call_sites, @@ -273,8 +323,26 @@ static struct notifier_block static_call_module_nb = { .notifier_call = static_call_module_notify, }; +#else + +static inline int __static_call_mod_text_reserved(void *start, void *end) +{ + return 0; +} + #endif /* CONFIG_MODULES */ +int static_call_text_reserved(void *start, void *end) +{ + int ret = __static_call_text_reserved(__start_static_call_sites, + __stop_static_call_sites, start, end); + + if (ret) + return ret; + + return __static_call_mod_text_reserved(start, end); +} + static void __init static_call_init(void) { int ret; -- cgit v1.2.3 From f03c412915f5f69f2d17bcd20ecdd69320bcbf7b Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 18 Aug 2020 15:57:46 +0200 Subject: static_call: Add simple self-test for static calls Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20200818135804.922581202@infradead.org --- kernel/static_call.c | 43 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 43 insertions(+) (limited to 'kernel/static_call.c') diff --git a/kernel/static_call.c b/kernel/static_call.c index 753b2f1b4fb8..97142cb6bfa6 100644 --- a/kernel/static_call.c +++ b/kernel/static_call.c @@ -369,3 +369,46 @@ static void __init static_call_init(void) #endif } early_initcall(static_call_init); + +#ifdef CONFIG_STATIC_CALL_SELFTEST + +static int func_a(int x) +{ + return x+1; +} + +static int func_b(int x) +{ + return x+2; +} + +DEFINE_STATIC_CALL(sc_selftest, func_a); + +static struct static_call_data { + int (*func)(int); + int val; + int expect; +} static_call_data [] __initdata = { + { NULL, 2, 3 }, + { func_b, 2, 4 }, + { func_a, 2, 3 } +}; + +static int __init test_static_call_init(void) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(static_call_data); i++ ) { + struct static_call_data *scd = &static_call_data[i]; + + if (scd->func) + static_call_update(sc_selftest, scd->func); + + WARN_ON(static_call(sc_selftest)(scd->val) != scd->expect); + } + + return 0; +} +early_initcall(test_static_call_init); + +#endif /* CONFIG_STATIC_CALL_SELFTEST */ -- cgit v1.2.3 From 5b06fd3bb9cdce4f3e731c48eb5b74c4acc47997 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 18 Aug 2020 15:57:49 +0200 Subject: static_call: Handle tail-calls GCC can turn our static_call(name)(args...) into a tail call, in which case we get a JMP.d32 into the trampoline (which then does a further tail-call). Teach objtool to recognise and mark these in .static_call_sites and adjust the code patching to deal with this. Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Cc: Linus Torvalds Link: https://lore.kernel.org/r/20200818135805.101186767@infradead.org --- kernel/static_call.c | 21 +++++++++++++-------- 1 file changed, 13 insertions(+), 8 deletions(-) (limited to 'kernel/static_call.c') diff --git a/kernel/static_call.c b/kernel/static_call.c index 97142cb6bfa6..d98e0e4272c1 100644 --- a/kernel/static_call.c +++ b/kernel/static_call.c @@ -15,8 +15,6 @@ extern struct static_call_site __start_static_call_sites[], static bool static_call_initialized; -#define STATIC_CALL_INIT 1UL - /* mutex to protect key modules/sites */ static DEFINE_MUTEX(static_call_mutex); @@ -39,18 +37,23 @@ static inline void *static_call_addr(struct static_call_site *site) static inline struct static_call_key *static_call_key(const struct static_call_site *site) { return (struct static_call_key *) - (((long)site->key + (long)&site->key) & ~STATIC_CALL_INIT); + (((long)site->key + (long)&site->key) & ~STATIC_CALL_SITE_FLAGS); } /* These assume the key is word-aligned. */ static inline bool static_call_is_init(struct static_call_site *site) { - return ((long)site->key + (long)&site->key) & STATIC_CALL_INIT; + return ((long)site->key + (long)&site->key) & STATIC_CALL_SITE_INIT; +} + +static inline bool static_call_is_tail(struct static_call_site *site) +{ + return ((long)site->key + (long)&site->key) & STATIC_CALL_SITE_TAIL; } static inline void static_call_set_init(struct static_call_site *site) { - site->key = ((long)static_call_key(site) | STATIC_CALL_INIT) - + site->key = ((long)static_call_key(site) | STATIC_CALL_SITE_INIT) - (long)&site->key; } @@ -104,7 +107,7 @@ void __static_call_update(struct static_call_key *key, void *tramp, void *func) key->func = func; - arch_static_call_transform(NULL, tramp, func); + arch_static_call_transform(NULL, tramp, func, false); /* * If uninitialized, we'll not update the callsites, but they still @@ -154,7 +157,8 @@ void __static_call_update(struct static_call_key *key, void *tramp, void *func) continue; } - arch_static_call_transform(site_addr, NULL, func); + arch_static_call_transform(site_addr, NULL, func, + static_call_is_tail(site)); } } @@ -198,7 +202,8 @@ static int __static_call_init(struct module *mod, key->mods = site_mod; } - arch_static_call_transform(site_addr, NULL, key->func); + arch_static_call_transform(site_addr, NULL, key->func, + static_call_is_tail(site)); } return 0; -- cgit v1.2.3 From a945c8345ec0decb2f1a7f19a8c5e60bcb1dd1eb Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 18 Aug 2020 15:57:51 +0200 Subject: static_call: Allow early init In order to use static_call() to wire up x86_pmu, we need to initialize earlier, specifically before memory allocation works; copy some of the tricks from jump_label to enable this. Primarily we overload key->next to store a sites pointer when there are no modules, this avoids having to use kmalloc() to initialize the sites and allows us to run much earlier. Signed-off-by: Peter Zijlstra (Intel) Signed-off-by: Ingo Molnar Reviewed-by: Steven Rostedt (VMware) Link: https://lore.kernel.org/r/20200818135805.220737930@infradead.org --- kernel/static_call.c | 70 +++++++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 66 insertions(+), 4 deletions(-) (limited to 'kernel/static_call.c') diff --git a/kernel/static_call.c b/kernel/static_call.c index d98e0e4272c1..f8362b3f8fd5 100644 --- a/kernel/static_call.c +++ b/kernel/static_call.c @@ -94,10 +94,31 @@ static inline void static_call_sort_entries(struct static_call_site *start, static_call_site_cmp, static_call_site_swap); } +static inline bool static_call_key_has_mods(struct static_call_key *key) +{ + return !(key->type & 1); +} + +static inline struct static_call_mod *static_call_key_next(struct static_call_key *key) +{ + if (!static_call_key_has_mods(key)) + return NULL; + + return key->mods; +} + +static inline struct static_call_site *static_call_key_sites(struct static_call_key *key) +{ + if (static_call_key_has_mods(key)) + return NULL; + + return (struct static_call_site *)(key->type & ~1); +} + void __static_call_update(struct static_call_key *key, void *tramp, void *func) { struct static_call_site *site, *stop; - struct static_call_mod *site_mod; + struct static_call_mod *site_mod, first; cpus_read_lock(); static_call_lock(); @@ -116,13 +137,22 @@ void __static_call_update(struct static_call_key *key, void *tramp, void *func) if (WARN_ON_ONCE(!static_call_initialized)) goto done; - for (site_mod = key->mods; site_mod; site_mod = site_mod->next) { + first = (struct static_call_mod){ + .next = static_call_key_next(key), + .mod = NULL, + .sites = static_call_key_sites(key), + }; + + for (site_mod = &first; site_mod; site_mod = site_mod->next) { struct module *mod = site_mod->mod; if (!site_mod->sites) { /* * This can happen if the static call key is defined in * a module which doesn't use it. + * + * It also happens in the has_mods case, where the + * 'first' entry has no sites associated with it. */ continue; } @@ -192,16 +222,48 @@ static int __static_call_init(struct module *mod, if (key != prev_key) { prev_key = key; + /* + * For vmlinux (!mod) avoid the allocation by storing + * the sites pointer in the key itself. Also see + * __static_call_update()'s @first. + * + * This allows architectures (eg. x86) to call + * static_call_init() before memory allocation works. + */ + if (!mod) { + key->sites = site; + key->type |= 1; + goto do_transform; + } + site_mod = kzalloc(sizeof(*site_mod), GFP_KERNEL); if (!site_mod) return -ENOMEM; + /* + * When the key has a direct sites pointer, extract + * that into an explicit struct static_call_mod, so we + * can have a list of modules. + */ + if (static_call_key_sites(key)) { + site_mod->mod = NULL; + site_mod->next = NULL; + site_mod->sites = static_call_key_sites(key); + + key->mods = site_mod; + + site_mod = kzalloc(sizeof(*site_mod), GFP_KERNEL); + if (!site_mod) + return -ENOMEM; + } + site_mod->mod = mod; site_mod->sites = site; - site_mod->next = key->mods; + site_mod->next = static_call_key_next(key); key->mods = site_mod; } +do_transform: arch_static_call_transform(site_addr, NULL, key->func, static_call_is_tail(site)); } @@ -348,7 +410,7 @@ int static_call_text_reserved(void *start, void *end) return __static_call_mod_text_reserved(start, end); } -static void __init static_call_init(void) +void __init static_call_init(void) { int ret; -- cgit v1.2.3 From 69e0ad37c9f32d5aa1beb02aab4ec0cd055be013 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Mon, 28 Sep 2020 16:09:39 -0700 Subject: static_call: Fix return type of static_call_init Functions that are passed to early_initcall should be of type initcall_t, which expects a return type of int. This is not currently an error but a patch in the Clang LTO series could change that in the future. Fixes: 9183c3f9ed71 ("static_call: Add inline static call infrastructure") Signed-off-by: Nathan Chancellor Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Sami Tolvanen Link: https://lore.kernel.org/lkml/20200903203053.3411268-17-samitolvanen@google.com/ --- kernel/static_call.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'kernel/static_call.c') diff --git a/kernel/static_call.c b/kernel/static_call.c index f8362b3f8fd5..84565c2a41b8 100644 --- a/kernel/static_call.c +++ b/kernel/static_call.c @@ -410,12 +410,12 @@ int static_call_text_reserved(void *start, void *end) return __static_call_mod_text_reserved(start, end); } -void __init static_call_init(void) +int __init static_call_init(void) { int ret; if (static_call_initialized) - return; + return 0; cpus_read_lock(); static_call_lock(); @@ -434,6 +434,7 @@ void __init static_call_init(void) #ifdef CONFIG_MODULES register_module_notifier(&static_call_module_nb); #endif + return 0; } early_initcall(static_call_init); -- cgit v1.2.3