Merge branch 'for-linus' of git://git.armlinux.org.uk/~rmk/linux-arm

Pull ARM updates from Russell King:
 "Lots of little things this time:

   - allow modules to be autoloaded according to the HWCAP feature bits
     (used primarily for crypto modules)

   - split module core and init PLT sections, since the core code and
     init code could be placed far apart, and the PLT sections need to
     be local to the code block.

   - three patches from Chris Brandt to allow Cortex-A9 L2 cache
     optimisations to be disabled where a SoC didn't wire up the out of
     band signals.

   - NoMMU compliance fixes, avoiding corruption of vector table which
     is not being used at this point, and avoiding possible register
     state corruption when switching mode.

   - fixmap memory attribute compliance update.

   - remove unnecessary locking from update_sections_early()

   - ftrace fix for DEBUG_RODATA with !FRAME_POINTER"

* 'for-linus' of git://git.armlinux.org.uk/~rmk/linux-arm:
  ARM: 8672/1: mm: remove tasklist locking from update_sections_early()
  ARM: 8671/1: V7M: Preserve registers across switch from Thread to Handler mode
  ARM: 8670/1: V7M: Do not corrupt vector table around v7m_invalidate_l1 call
  ARM: 8668/1: ftrace: Fix dynamic ftrace with DEBUG_RODATA and !FRAME_POINTER
  ARM: 8667/3: Fix memory attribute inconsistencies when using fixmap
  ARM: 8663/1: wire up HWCAP/HWCAP2 feature bits to the CPU modalias
  ARM: 8666/1: mm: dump: Add domain to output
  ARM: 8662/1: module: split core and init PLT sections
  ARM: 8661/1: dts: r7s72100: add l2 cache
  ARM: 8660/1: shmobile: r7s72100: Enable L2 cache
  ARM: 8659/1: l2c: allow CA9 optimizations to be disabled

4 files changed, 69 insertions, 34 deletions

diff --git a/arch/arm/kernel/ftrace.c b/arch/arm/kernel/ftrace.c
index 3f1759411d51..414e60ed0257 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;
diff --git a/arch/arm/kernel/module-plts.c b/arch/arm/kernel/module-plts.c
index 3a5cba90c971..3d0c2e4dda1d 100644
--- a/arch/arm/kernel/module-plts.c
+++ b/arch/arm/kernel/module-plts.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2014 Linaro Ltd. <ard.biesheuvel@linaro.org>
+ * Copyright (C) 2014-2017 Linaro Ltd. <ard.biesheuvel@linaro.org>
  *
  * 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 05881e2b414c..eacb5c67f61e 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) }
 }
diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index f4e54503afa9..32e1a9513dc7 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();