From 2d25bc55235314d869459c574be14e8faa73aca3 Mon Sep 17 00:00:00 2001 From: Jessica Yu Date: Mon, 19 Nov 2018 17:43:58 +0100 Subject: module: make it clearer when we're handling kallsyms symbols vs exported symbols The module loader internally works with both exported symbols represented as struct kernel_symbol, as well as Elf symbols from a module's symbol table. It's hard to distinguish sometimes which type of symbol we're handling given that some helper function names are not consistent or helpful. Take get_ksymbol() for instance - are we looking for an exported symbol or a kallsyms symbol here? Or symname() and kernel_symbol_name() - which function handles an exported symbol and which one an Elf symbol? Clean up and unify the function naming scheme a bit to make it clear which kind of symbol we're handling. This change only affects static functions internal to the module loader. Reviewed-by: Miroslav Benes Signed-off-by: Jessica Yu --- kernel/module.c | 78 ++++++++++++++++++++++++++++++++------------------------- 1 file changed, 44 insertions(+), 34 deletions(-) (limited to 'kernel/module.c') diff --git a/kernel/module.c b/kernel/module.c index 49a405891587..1b5edf78694c 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -495,9 +495,9 @@ struct find_symbol_arg { const struct kernel_symbol *sym; }; -static bool check_symbol(const struct symsearch *syms, - struct module *owner, - unsigned int symnum, void *data) +static bool check_exported_symbol(const struct symsearch *syms, + struct module *owner, + unsigned int symnum, void *data) { struct find_symbol_arg *fsa = data; @@ -555,9 +555,9 @@ static int cmp_name(const void *va, const void *vb) return strcmp(a, kernel_symbol_name(b)); } -static bool find_symbol_in_section(const struct symsearch *syms, - struct module *owner, - void *data) +static bool find_exported_symbol_in_section(const struct symsearch *syms, + struct module *owner, + void *data) { struct find_symbol_arg *fsa = data; struct kernel_symbol *sym; @@ -565,13 +565,14 @@ static bool find_symbol_in_section(const struct symsearch *syms, sym = bsearch(fsa->name, syms->start, syms->stop - syms->start, sizeof(struct kernel_symbol), cmp_name); - if (sym != NULL && check_symbol(syms, owner, sym - syms->start, data)) + if (sym != NULL && check_exported_symbol(syms, owner, + sym - syms->start, data)) return true; return false; } -/* Find a symbol and return it, along with, (optional) crc and +/* Find an exported symbol and return it, along with, (optional) crc and * (optional) module which owns it. Needs preempt disabled or module_mutex. */ const struct kernel_symbol *find_symbol(const char *name, struct module **owner, @@ -585,7 +586,7 @@ const struct kernel_symbol *find_symbol(const char *name, fsa.gplok = gplok; fsa.warn = warn; - if (each_symbol_section(find_symbol_in_section, &fsa)) { + if (each_symbol_section(find_exported_symbol_in_section, &fsa)) { if (owner) *owner = fsa.owner; if (crc) @@ -2198,7 +2199,7 @@ EXPORT_SYMBOL_GPL(__symbol_get); * * You must hold the module_mutex. */ -static int verify_export_symbols(struct module *mod) +static int verify_exported_symbols(struct module *mod) { unsigned int i; struct module *owner; @@ -2519,10 +2520,10 @@ static void free_modinfo(struct module *mod) #ifdef CONFIG_KALLSYMS -/* lookup symbol in given range of kernel_symbols */ -static const struct kernel_symbol *lookup_symbol(const char *name, - const struct kernel_symbol *start, - const struct kernel_symbol *stop) +/* Lookup exported symbol in given range of kernel_symbols */ +static const struct kernel_symbol *lookup_exported_symbol(const char *name, + const struct kernel_symbol *start, + const struct kernel_symbol *stop) { return bsearch(name, start, stop - start, sizeof(struct kernel_symbol), cmp_name); @@ -2533,9 +2534,10 @@ static int is_exported(const char *name, unsigned long value, { const struct kernel_symbol *ks; if (!mod) - ks = lookup_symbol(name, __start___ksymtab, __stop___ksymtab); + ks = lookup_exported_symbol(name, __start___ksymtab, __stop___ksymtab); else - ks = lookup_symbol(name, mod->syms, mod->syms + mod->num_syms); + ks = lookup_exported_symbol(name, mod->syms, mod->syms + mod->num_syms); + return ks != NULL && kernel_symbol_value(ks) == value; } @@ -3592,7 +3594,7 @@ static int complete_formation(struct module *mod, struct load_info *info) mutex_lock(&module_mutex); /* Find duplicate symbols (must be called under lock). */ - err = verify_export_symbols(mod); + err = verify_exported_symbols(mod); if (err < 0) goto out; @@ -3911,15 +3913,19 @@ static inline int is_arm_mapping_symbol(const char *str) && (str[2] == '\0' || str[2] == '.'); } -static const char *symname(struct mod_kallsyms *kallsyms, unsigned int symnum) +static const char *kallsyms_symbol_name(struct mod_kallsyms *kallsyms, unsigned int symnum) { return kallsyms->strtab + kallsyms->symtab[symnum].st_name; } -static const char *get_ksymbol(struct module *mod, - unsigned long addr, - unsigned long *size, - unsigned long *offset) +/* + * Given a module and address, find the corresponding symbol and return its name + * while providing its size and offset if needed. + */ +static const char *find_kallsyms_symbol(struct module *mod, + unsigned long addr, + unsigned long *size, + unsigned long *offset) { unsigned int i, best = 0; unsigned long nextval; @@ -3939,8 +3945,8 @@ static const char *get_ksymbol(struct module *mod, /* We ignore unnamed symbols: they're uninformative * and inserted at a whim. */ - if (*symname(kallsyms, i) == '\0' - || is_arm_mapping_symbol(symname(kallsyms, i))) + if (*kallsyms_symbol_name(kallsyms, i) == '\0' + || is_arm_mapping_symbol(kallsyms_symbol_name(kallsyms, i))) continue; if (kallsyms->symtab[i].st_value <= addr @@ -3958,7 +3964,8 @@ static const char *get_ksymbol(struct module *mod, *size = nextval - kallsyms->symtab[best].st_value; if (offset) *offset = addr - kallsyms->symtab[best].st_value; - return symname(kallsyms, best); + + return kallsyms_symbol_name(kallsyms, best); } void * __weak dereference_module_function_descriptor(struct module *mod, @@ -3983,7 +3990,8 @@ const char *module_address_lookup(unsigned long addr, if (mod) { if (modname) *modname = mod->name; - ret = get_ksymbol(mod, addr, size, offset); + + ret = find_kallsyms_symbol(mod, addr, size, offset); } /* Make a copy in here where it's safe */ if (ret) { @@ -4006,9 +4014,10 @@ int lookup_module_symbol_name(unsigned long addr, char *symname) if (within_module(addr, mod)) { const char *sym; - sym = get_ksymbol(mod, addr, NULL, NULL); + sym = find_kallsyms_symbol(mod, addr, NULL, NULL); if (!sym) goto out; + strlcpy(symname, sym, KSYM_NAME_LEN); preempt_enable(); return 0; @@ -4031,7 +4040,7 @@ int lookup_module_symbol_attrs(unsigned long addr, unsigned long *size, if (within_module(addr, mod)) { const char *sym; - sym = get_ksymbol(mod, addr, size, offset); + sym = find_kallsyms_symbol(mod, addr, size, offset); if (!sym) goto out; if (modname) @@ -4062,7 +4071,7 @@ int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type, if (symnum < kallsyms->num_symtab) { *value = kallsyms->symtab[symnum].st_value; *type = kallsyms->symtab[symnum].st_info; - strlcpy(name, symname(kallsyms, symnum), KSYM_NAME_LEN); + strlcpy(name, kallsyms_symbol_name(kallsyms, symnum), KSYM_NAME_LEN); strlcpy(module_name, mod->name, MODULE_NAME_LEN); *exported = is_exported(name, *value, mod); preempt_enable(); @@ -4074,13 +4083,14 @@ int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type, return -ERANGE; } -static unsigned long mod_find_symname(struct module *mod, const char *name) +/* Given a module and name of symbol, find and return the symbol's value */ +static unsigned long find_kallsyms_symbol_value(struct module *mod, const char *name) { unsigned int i; struct mod_kallsyms *kallsyms = rcu_dereference_sched(mod->kallsyms); for (i = 0; i < kallsyms->num_symtab; i++) - if (strcmp(name, symname(kallsyms, i)) == 0 && + if (strcmp(name, kallsyms_symbol_name(kallsyms, i)) == 0 && kallsyms->symtab[i].st_shndx != SHN_UNDEF) return kallsyms->symtab[i].st_value; return 0; @@ -4097,12 +4107,12 @@ unsigned long module_kallsyms_lookup_name(const char *name) preempt_disable(); if ((colon = strnchr(name, MODULE_NAME_LEN, ':')) != NULL) { if ((mod = find_module_all(name, colon - name, false)) != NULL) - ret = mod_find_symname(mod, colon+1); + ret = find_kallsyms_symbol_value(mod, colon+1); } else { list_for_each_entry_rcu(mod, &modules, list) { if (mod->state == MODULE_STATE_UNFORMED) continue; - if ((ret = mod_find_symname(mod, name)) != 0) + if ((ret = find_kallsyms_symbol_value(mod, name)) != 0) break; } } @@ -4131,7 +4141,7 @@ int module_kallsyms_on_each_symbol(int (*fn)(void *, const char *, if (kallsyms->symtab[i].st_shndx == SHN_UNDEF) continue; - ret = fn(data, symname(kallsyms, i), + ret = fn(data, kallsyms_symbol_name(kallsyms, i), mod, kallsyms->symtab[i].st_value); if (ret != 0) return ret; -- cgit v1.2.3-58-ga151 From 5439c985c5a83a8419f762115afdf560ab72a452 Mon Sep 17 00:00:00 2001 From: Vincent Whitchurch Date: Fri, 14 Dec 2018 17:05:54 +0100 Subject: module: Overwrite st_size instead of st_info st_info is currently overwritten after relocation and used to store the elf_type(). However, we're going to need it fix kallsyms on ARM's Thumb-2 kernels, so preserve st_info and overwrite the st_size field instead. st_size is neither used by the module core nor by any architecture. Reviewed-by: Miroslav Benes Reviewed-by: Dave Martin Signed-off-by: Vincent Whitchurch Signed-off-by: Jessica Yu --- kernel/module.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/module.c') diff --git a/kernel/module.c b/kernel/module.c index 1b5edf78694c..b36ff8a3d562 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -2684,7 +2684,7 @@ static void add_kallsyms(struct module *mod, const struct load_info *info) /* Set types up while we still have access to sections. */ for (i = 0; i < mod->kallsyms->num_symtab; i++) - mod->kallsyms->symtab[i].st_info + mod->kallsyms->symtab[i].st_size = elf_type(&mod->kallsyms->symtab[i], info); /* Now populate the cut down core kallsyms for after init. */ @@ -4070,7 +4070,7 @@ int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type, kallsyms = rcu_dereference_sched(mod->kallsyms); if (symnum < kallsyms->num_symtab) { *value = kallsyms->symtab[symnum].st_value; - *type = kallsyms->symtab[symnum].st_info; + *type = kallsyms->symtab[symnum].st_size; strlcpy(name, kallsyms_symbol_name(kallsyms, symnum), KSYM_NAME_LEN); strlcpy(module_name, mod->name, MODULE_NAME_LEN); *exported = is_exported(name, *value, mod); -- cgit v1.2.3-58-ga151 From 93d77e7f1410c366050d6035dcba1a5167c7cf0b Mon Sep 17 00:00:00 2001 From: Vincent Whitchurch Date: Fri, 14 Dec 2018 17:05:55 +0100 Subject: ARM: module: Fix function kallsyms on Thumb-2 Thumb-2 functions have the lowest bit set in the symbol value in the symtab. When kallsyms are generated for the vmlinux, the kallsyms are generated from the output of nm, and nm clears the lowest bit. $ arm-linux-gnueabihf-readelf -a vmlinux | grep show_interrupts 95947: 8015dc89 686 FUNC GLOBAL DEFAULT 2 show_interrupts $ arm-linux-gnueabihf-nm vmlinux | grep show_interrupts 8015dc88 T show_interrupts $ cat /proc/kallsyms | grep show_interrupts 8015dc88 T show_interrupts However, for modules, the kallsyms uses the values in the symbol table without modification, so for functions in modules, the lowest bit is set in kallsyms. $ arm-linux-gnueabihf-readelf -a drivers/net/tun.ko | grep tun_get_socket 333: 00002d4d 36 FUNC GLOBAL DEFAULT 1 tun_get_socket $ arm-linux-gnueabihf-nm drivers/net/tun.ko | grep tun_get_socket 00002d4c T tun_get_socket $ cat /proc/kallsyms | grep tun_get_socket 7f802d4d t tun_get_socket [tun] Because of this, the symbol+offset of the crashing instruction shown in oopses is incorrect when the crash is in a module. For example, given a tun_get_socket which starts like this, 00002d4c : 2d4c: 6943 ldr r3, [r0, #20] 2d4e: 4a07 ldr r2, [pc, #28] 2d50: 4293 cmp r3, r2 a crash when tun_get_socket is called with NULL results in: PC is at tun_xdp+0xa3/0xa4 [tun] pc : [<7f802d4c>] As can be seen, the "PC is at" line reports the wrong symbol name, and the symbol+offset will point to the wrong source line if it is passed to gdb. To solve this, add a way for archs to fixup the reading of these module kallsyms values, and use that to clear the lowest bit for function symbols on Thumb-2. After the fix: # cat /proc/kallsyms | grep tun_get_socket 7f802d4c t tun_get_socket [tun] PC is at tun_get_socket+0x0/0x24 [tun] pc : [<7f802d4c>] Signed-off-by: Vincent Whitchurch Signed-off-by: Jessica Yu --- arch/arm/include/asm/module.h | 11 +++++++++++ include/linux/module.h | 7 +++++++ kernel/module.c | 43 +++++++++++++++++++++++++++---------------- 3 files changed, 45 insertions(+), 16 deletions(-) (limited to 'kernel/module.c') diff --git a/arch/arm/include/asm/module.h b/arch/arm/include/asm/module.h index 9e81b7c498d8..182163b55546 100644 --- a/arch/arm/include/asm/module.h +++ b/arch/arm/include/asm/module.h @@ -61,4 +61,15 @@ u32 get_module_plt(struct module *mod, unsigned long loc, Elf32_Addr val); MODULE_ARCH_VERMAGIC_ARMTHUMB \ MODULE_ARCH_VERMAGIC_P2V +#ifdef CONFIG_THUMB2_KERNEL +#define HAVE_ARCH_KALLSYMS_SYMBOL_VALUE +static inline unsigned long kallsyms_symbol_value(const Elf_Sym *sym) +{ + if (ELF_ST_TYPE(sym->st_info) == STT_FUNC) + return sym->st_value & ~1; + + return sym->st_value; +} +#endif + #endif /* _ASM_ARM_MODULE_H */ diff --git a/include/linux/module.h b/include/linux/module.h index fce6b4335e36..c0b4b7840b57 100644 --- a/include/linux/module.h +++ b/include/linux/module.h @@ -486,6 +486,13 @@ struct module { #define MODULE_ARCH_INIT {} #endif +#ifndef HAVE_ARCH_KALLSYMS_SYMBOL_VALUE +static inline unsigned long kallsyms_symbol_value(const Elf_Sym *sym) +{ + return sym->st_value; +} +#endif + extern struct mutex module_mutex; /* FIXME: It'd be nice to isolate modules during init, too, so they diff --git a/kernel/module.c b/kernel/module.c index b36ff8a3d562..164bf201eae4 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -3928,7 +3928,7 @@ static const char *find_kallsyms_symbol(struct module *mod, unsigned long *offset) { unsigned int i, best = 0; - unsigned long nextval; + unsigned long nextval, bestval; struct mod_kallsyms *kallsyms = rcu_dereference_sched(mod->kallsyms); /* At worse, next value is at end of module */ @@ -3937,10 +3937,15 @@ static const char *find_kallsyms_symbol(struct module *mod, else nextval = (unsigned long)mod->core_layout.base+mod->core_layout.text_size; + bestval = kallsyms_symbol_value(&kallsyms->symtab[best]); + /* Scan for closest preceding symbol, and next symbol. (ELF starts real symbols at 1). */ for (i = 1; i < kallsyms->num_symtab; i++) { - if (kallsyms->symtab[i].st_shndx == SHN_UNDEF) + const Elf_Sym *sym = &kallsyms->symtab[i]; + unsigned long thisval = kallsyms_symbol_value(sym); + + if (sym->st_shndx == SHN_UNDEF) continue; /* We ignore unnamed symbols: they're uninformative @@ -3949,21 +3954,21 @@ static const char *find_kallsyms_symbol(struct module *mod, || is_arm_mapping_symbol(kallsyms_symbol_name(kallsyms, i))) continue; - if (kallsyms->symtab[i].st_value <= addr - && kallsyms->symtab[i].st_value > kallsyms->symtab[best].st_value) + if (thisval <= addr && thisval > bestval) { best = i; - if (kallsyms->symtab[i].st_value > addr - && kallsyms->symtab[i].st_value < nextval) - nextval = kallsyms->symtab[i].st_value; + bestval = thisval; + } + if (thisval > addr && thisval < nextval) + nextval = thisval; } if (!best) return NULL; if (size) - *size = nextval - kallsyms->symtab[best].st_value; + *size = nextval - bestval; if (offset) - *offset = addr - kallsyms->symtab[best].st_value; + *offset = addr - bestval; return kallsyms_symbol_name(kallsyms, best); } @@ -4069,8 +4074,10 @@ int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type, continue; kallsyms = rcu_dereference_sched(mod->kallsyms); if (symnum < kallsyms->num_symtab) { - *value = kallsyms->symtab[symnum].st_value; - *type = kallsyms->symtab[symnum].st_size; + const Elf_Sym *sym = &kallsyms->symtab[symnum]; + + *value = kallsyms_symbol_value(sym); + *type = sym->st_size; strlcpy(name, kallsyms_symbol_name(kallsyms, symnum), KSYM_NAME_LEN); strlcpy(module_name, mod->name, MODULE_NAME_LEN); *exported = is_exported(name, *value, mod); @@ -4089,10 +4096,13 @@ static unsigned long find_kallsyms_symbol_value(struct module *mod, const char * unsigned int i; struct mod_kallsyms *kallsyms = rcu_dereference_sched(mod->kallsyms); - for (i = 0; i < kallsyms->num_symtab; i++) + for (i = 0; i < kallsyms->num_symtab; i++) { + const Elf_Sym *sym = &kallsyms->symtab[i]; + if (strcmp(name, kallsyms_symbol_name(kallsyms, i)) == 0 && - kallsyms->symtab[i].st_shndx != SHN_UNDEF) - return kallsyms->symtab[i].st_value; + sym->st_shndx != SHN_UNDEF) + return kallsyms_symbol_value(sym); + } return 0; } @@ -4137,12 +4147,13 @@ int module_kallsyms_on_each_symbol(int (*fn)(void *, const char *, if (mod->state == MODULE_STATE_UNFORMED) continue; for (i = 0; i < kallsyms->num_symtab; i++) { + const Elf_Sym *sym = &kallsyms->symtab[i]; - if (kallsyms->symtab[i].st_shndx == SHN_UNDEF) + if (sym->st_shndx == SHN_UNDEF) continue; ret = fn(data, kallsyms_symbol_name(kallsyms, i), - mod, kallsyms->symtab[i].st_value); + mod, kallsyms_symbol_value(sym)); if (ret != 0) return ret; } -- cgit v1.2.3-58-ga151