From b7ede5a1f5905ac394cc8e61712a13e3c5cb7b8f Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Wed, 22 Feb 2017 19:40:12 +0100 Subject: ARM: 8662/1: module: split core and init PLT sections Since commit 35fa91eed817 ("ARM: kernel: merge core and init PLTs"), the ARM module PLT code allocates all PLT entries in a single core section, since the overhead of having a separate init PLT section is not justified by the small number of PLT entries usually required for init code. However, the core and init module regions are allocated independently, and there is a corner case where the core region may be allocated from the VMALLOC region if the dedicated module region is exhausted, but the init region, being much smaller, can still be allocated from the module region. This puts the PLT entries out of reach of the relocated branch instructions, defeating the whole purpose of PLTs. So split the core and init PLT regions, and name the latter ".init.plt" so it gets allocated along with (and sufficiently close to) the .init sections that it serves. Also, given that init PLT entries may need to be emitted for branches that target the core module, modify the logic that disregards defined symbols to only disregard symbols that are defined in the same section. Fixes: 35fa91eed817 ("ARM: kernel: merge core and init PLTs") Cc: # v4.9+ Reported-by: Angus Clark Tested-by: Angus Clark Signed-off-by: Ard Biesheuvel Signed-off-by: Russell King --- arch/arm/kernel/module-plts.c | 87 +++++++++++++++++++++++++++++-------------- arch/arm/kernel/module.lds | 1 + 2 files changed, 61 insertions(+), 27 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/module-plts.c b/arch/arm/kernel/module-plts.c index 3a5cba9..3d0c2e4 100644 --- a/arch/arm/kernel/module-plts.c +++ b/arch/arm/kernel/module-plts.c @@ -1,5 +1,5 @@ /* - * Copyright (C) 2014 Linaro Ltd. + * Copyright (C) 2014-2017 Linaro Ltd. * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License version 2 as @@ -31,9 +31,17 @@ struct plt_entries { u32 lit[PLT_ENT_COUNT]; }; +static bool in_init(const struct module *mod, unsigned long loc) +{ + return loc - (u32)mod->init_layout.base < mod->init_layout.size; +} + u32 get_module_plt(struct module *mod, unsigned long loc, Elf32_Addr val) { - struct plt_entries *plt = (struct plt_entries *)mod->arch.plt->sh_addr; + struct mod_plt_sec *pltsec = !in_init(mod, loc) ? &mod->arch.core : + &mod->arch.init; + + struct plt_entries *plt = (struct plt_entries *)pltsec->plt->sh_addr; int idx = 0; /* @@ -41,9 +49,9 @@ u32 get_module_plt(struct module *mod, unsigned long loc, Elf32_Addr val) * relocations are sorted, this will be the last entry we allocated. * (if one exists). */ - if (mod->arch.plt_count > 0) { - plt += (mod->arch.plt_count - 1) / PLT_ENT_COUNT; - idx = (mod->arch.plt_count - 1) % PLT_ENT_COUNT; + if (pltsec->plt_count > 0) { + plt += (pltsec->plt_count - 1) / PLT_ENT_COUNT; + idx = (pltsec->plt_count - 1) % PLT_ENT_COUNT; if (plt->lit[idx] == val) return (u32)&plt->ldr[idx]; @@ -53,8 +61,8 @@ u32 get_module_plt(struct module *mod, unsigned long loc, Elf32_Addr val) plt++; } - mod->arch.plt_count++; - BUG_ON(mod->arch.plt_count * PLT_ENT_SIZE > mod->arch.plt->sh_size); + pltsec->plt_count++; + BUG_ON(pltsec->plt_count * PLT_ENT_SIZE > pltsec->plt->sh_size); if (!idx) /* Populate a new set of entries */ @@ -129,7 +137,7 @@ static bool duplicate_rel(Elf32_Addr base, const Elf32_Rel *rel, int num) /* Count how many PLT entries we may need */ static unsigned int count_plts(const Elf32_Sym *syms, Elf32_Addr base, - const Elf32_Rel *rel, int num) + const Elf32_Rel *rel, int num, Elf32_Word dstidx) { unsigned int ret = 0; const Elf32_Sym *s; @@ -144,13 +152,17 @@ static unsigned int count_plts(const Elf32_Sym *syms, Elf32_Addr base, case R_ARM_THM_JUMP24: /* * We only have to consider branch targets that resolve - * to undefined symbols. This is not simply a heuristic, - * it is a fundamental limitation, since the PLT itself - * is part of the module, and needs to be within range - * as well, so modules can never grow beyond that limit. + * to symbols that are defined in a different section. + * This is not simply a heuristic, it is a fundamental + * limitation, since there is no guaranteed way to emit + * PLT entries sufficiently close to the branch if the + * section size exceeds the range of a branch + * instruction. So ignore relocations against defined + * symbols if they live in the same section as the + * relocation target. */ s = syms + ELF32_R_SYM(rel[i].r_info); - if (s->st_shndx != SHN_UNDEF) + if (s->st_shndx == dstidx) break; /* @@ -161,7 +173,12 @@ static unsigned int count_plts(const Elf32_Sym *syms, Elf32_Addr base, * So we need to support them, but there is no need to * take them into consideration when trying to optimize * this code. So let's only check for duplicates when - * the addend is zero. + * the addend is zero. (Note that calls into the core + * module via init PLT entries could involve section + * relative symbol references with non-zero addends, for + * which we may end up emitting duplicates, but the init + * PLT is released along with the rest of the .init + * region as soon as module loading completes.) */ if (!is_zero_addend_relocation(base, rel + i) || !duplicate_rel(base, rel, i)) @@ -174,7 +191,8 @@ static unsigned int count_plts(const Elf32_Sym *syms, Elf32_Addr base, int module_frob_arch_sections(Elf_Ehdr *ehdr, Elf_Shdr *sechdrs, char *secstrings, struct module *mod) { - unsigned long plts = 0; + unsigned long core_plts = 0; + unsigned long init_plts = 0; Elf32_Shdr *s, *sechdrs_end = sechdrs + ehdr->e_shnum; Elf32_Sym *syms = NULL; @@ -184,13 +202,15 @@ int module_frob_arch_sections(Elf_Ehdr *ehdr, Elf_Shdr *sechdrs, */ for (s = sechdrs; s < sechdrs_end; ++s) { if (strcmp(".plt", secstrings + s->sh_name) == 0) - mod->arch.plt = s; + mod->arch.core.plt = s; + else if (strcmp(".init.plt", secstrings + s->sh_name) == 0) + mod->arch.init.plt = s; else if (s->sh_type == SHT_SYMTAB) syms = (Elf32_Sym *)s->sh_addr; } - if (!mod->arch.plt) { - pr_err("%s: module PLT section missing\n", mod->name); + if (!mod->arch.core.plt || !mod->arch.init.plt) { + pr_err("%s: module PLT section(s) missing\n", mod->name); return -ENOEXEC; } if (!syms) { @@ -213,16 +233,29 @@ int module_frob_arch_sections(Elf_Ehdr *ehdr, Elf_Shdr *sechdrs, /* sort by type and symbol index */ sort(rels, numrels, sizeof(Elf32_Rel), cmp_rel, NULL); - plts += count_plts(syms, dstsec->sh_addr, rels, numrels); + if (strncmp(secstrings + dstsec->sh_name, ".init", 5) != 0) + core_plts += count_plts(syms, dstsec->sh_addr, rels, + numrels, s->sh_info); + else + init_plts += count_plts(syms, dstsec->sh_addr, rels, + numrels, s->sh_info); } - mod->arch.plt->sh_type = SHT_NOBITS; - mod->arch.plt->sh_flags = SHF_EXECINSTR | SHF_ALLOC; - mod->arch.plt->sh_addralign = L1_CACHE_BYTES; - mod->arch.plt->sh_size = round_up(plts * PLT_ENT_SIZE, - sizeof(struct plt_entries)); - mod->arch.plt_count = 0; - - pr_debug("%s: plt=%x\n", __func__, mod->arch.plt->sh_size); + mod->arch.core.plt->sh_type = SHT_NOBITS; + mod->arch.core.plt->sh_flags = SHF_EXECINSTR | SHF_ALLOC; + mod->arch.core.plt->sh_addralign = L1_CACHE_BYTES; + mod->arch.core.plt->sh_size = round_up(core_plts * PLT_ENT_SIZE, + sizeof(struct plt_entries)); + mod->arch.core.plt_count = 0; + + mod->arch.init.plt->sh_type = SHT_NOBITS; + mod->arch.init.plt->sh_flags = SHF_EXECINSTR | SHF_ALLOC; + mod->arch.init.plt->sh_addralign = L1_CACHE_BYTES; + mod->arch.init.plt->sh_size = round_up(init_plts * PLT_ENT_SIZE, + sizeof(struct plt_entries)); + mod->arch.init.plt_count = 0; + + pr_debug("%s: plt=%x, init.plt=%x\n", __func__, + mod->arch.core.plt->sh_size, mod->arch.init.plt->sh_size); return 0; } diff --git a/arch/arm/kernel/module.lds b/arch/arm/kernel/module.lds index 05881e2..eacb5c6 100644 --- a/arch/arm/kernel/module.lds +++ b/arch/arm/kernel/module.lds @@ -1,3 +1,4 @@ SECTIONS { .plt : { BYTE(0) } + .init.plt : { BYTE(0) } } -- cgit v1.1 From b089c31c519c3906c14801b6ec483e18a5152a50 Mon Sep 17 00:00:00 2001 From: Jon Medhurst Date: Mon, 10 Apr 2017 11:13:59 +0100 Subject: ARM: 8667/3: Fix memory attribute inconsistencies when using fixmap To cope with the variety in ARM architectures and configurations, the pagetable attributes for kernel memory are generated at runtime to match the system the kernel finds itself on. This calculated value is stored in pgprot_kernel. However, when early fixmap support was added for ARM (commit a5f4c561b3b1) the attributes used for mappings were hard coded because pgprot_kernel is not set up early enough. Unfortunately, when fixmap is used after early boot this means the memory being mapped can have different attributes to existing mappings, potentially leading to unpredictable behaviour. A specific problem also exists due to the hard coded values not include the 'shareable' attribute which means on systems where this matters (e.g. those with multiple CPU clusters) the cache contents for a memory location can become inconsistent between CPUs. To resolve these issues we change fixmap to use the same memory attributes (from pgprot_kernel) that the rest of the kernel uses. To enable this we need to refactor the initialisation code so build_mem_type_table() is called early enough. Note, that relies on early param parsing for memory type overrides passed via the kernel command line, so we need to make sure this call is still after parse_early_params(). [ardb: keep early_fixmap_init() before param parsing, for earlycon] Fixes: a5f4c561b3b1 ("ARM: 8415/1: early fixmap support for earlycon") Cc: # v4.3+ Tested-by: afzal mohammed Signed-off-by: Jon Medhurst Signed-off-by: Ard Biesheuvel Signed-off-by: Russell King --- arch/arm/kernel/setup.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c index f4e5450..32e1a95 100644 --- a/arch/arm/kernel/setup.c +++ b/arch/arm/kernel/setup.c @@ -80,7 +80,7 @@ __setup("fpe=", fpe_setup); extern void init_default_cache_policy(unsigned long); extern void paging_init(const struct machine_desc *desc); -extern void early_paging_init(const struct machine_desc *); +extern void early_mm_init(const struct machine_desc *); extern void adjust_lowmem_bounds(void); extern enum reboot_mode reboot_mode; extern void setup_dma_zone(const struct machine_desc *desc); @@ -1088,7 +1088,7 @@ void __init setup_arch(char **cmdline_p) parse_early_param(); #ifdef CONFIG_MMU - early_paging_init(mdesc); + early_mm_init(mdesc); #endif setup_dma_zone(mdesc); xen_early_init(); -- cgit v1.1 From 6f05d0761af612e04572ba4d65b4c0274a88444f Mon Sep 17 00:00:00 2001 From: Abel Vesa Date: Mon, 3 Apr 2017 23:58:54 +0100 Subject: ARM: 8668/1: ftrace: Fix dynamic ftrace with DEBUG_RODATA and !FRAME_POINTER The support for dynamic ftrace with CONFIG_DEBUG_RODATA involves overriding the weak arch_ftrace_update_code() with a variant which makes the kernel text writable around the patching. This override was however added under the CONFIG_OLD_MCOUNT ifdef, and CONFIG_OLD_MCOUNT is only enabled if frame pointers are enabled. This leads to non-functional dynamic ftrace (ftrace triggers a WARN_ON()) when CONFIG_DEBUG_RODATA is enabled and CONFIG_FRAME_POINTER is not. Move the override out of that ifdef and into the CONFIG_DYNAMIC_FTRACE ifdef where it belongs. Fixes: 80d6b0c2eed2a ("ARM: mm: allow text and rodata sections to be read-only") Suggested-by: Nicolai Stange Suggested-by: Rabin Vincent Signed-off-by: Abel Vesa Acked-by: Rabin Vincent Signed-off-by: Russell King --- arch/arm/kernel/ftrace.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'arch/arm/kernel') diff --git a/arch/arm/kernel/ftrace.c b/arch/arm/kernel/ftrace.c index 3f17594..414e60e 100644 --- a/arch/arm/kernel/ftrace.c +++ b/arch/arm/kernel/ftrace.c @@ -29,11 +29,6 @@ #endif #ifdef CONFIG_DYNAMIC_FTRACE -#ifdef CONFIG_OLD_MCOUNT -#define OLD_MCOUNT_ADDR ((unsigned long) mcount) -#define OLD_FTRACE_ADDR ((unsigned long) ftrace_caller_old) - -#define OLD_NOP 0xe1a00000 /* mov r0, r0 */ static int __ftrace_modify_code(void *data) { @@ -51,6 +46,12 @@ void arch_ftrace_update_code(int command) stop_machine(__ftrace_modify_code, &command, NULL); } +#ifdef CONFIG_OLD_MCOUNT +#define OLD_MCOUNT_ADDR ((unsigned long) mcount) +#define OLD_FTRACE_ADDR ((unsigned long) ftrace_caller_old) + +#define OLD_NOP 0xe1a00000 /* mov r0, r0 */ + static unsigned long ftrace_nop_replace(struct dyn_ftrace *rec) { return rec->arch.old_mcount ? OLD_NOP : NOP; -- cgit v1.1