[PATCH v4 10/18] x86/static_call: Add inline static call implementation for x86-64

From: Peter Zijlstra
Date: Fri May 01 2020 - 16:31:54 EST


From: Josh Poimboeuf <jpoimboe@xxxxxxxxxx>

Add the inline static call implementation for x86-64. The generated code
is identical to the out-of-line case, except we move the trampoline into
it's own section.

Objtool uses the trampoline naming convention to detect all the call
sites. It then annotates those call sites in the .static_call_sites
section.

During boot (and module init), the call sites are patched to call
directly into the destination function. The temporary trampoline is
then no longer used.

[peterz: merged trampolines, put trampoline in section]
Signed-off-by: Josh Poimboeuf <jpoimboe@xxxxxxxxxx>
Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
---
arch/x86/Kconfig | 3
arch/x86/include/asm/static_call.h | 13 ++-
arch/x86/kernel/static_call.c | 5 -
arch/x86/kernel/vmlinux.lds.S | 1
include/asm-generic/vmlinux.lds.h | 6 +
tools/include/linux/static_call_types.h | 28 ++++++
tools/objtool/check.c | 130 +++++++++++++++++++++++++++++++-
tools/objtool/check.h | 2
tools/objtool/elf.c | 6 -
tools/objtool/elf.h | 3
tools/objtool/orc_gen.c | 4
tools/objtool/sync-check.sh | 1
12 files changed, 192 insertions(+), 10 deletions(-)
create mode 100644 tools/objtool/include/linux/static_call_types.h

--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -211,6 +211,7 @@ config X86
select HAVE_STACKPROTECTOR if CC_HAS_SANE_STACKPROTECTOR
select HAVE_STACK_VALIDATION if X86_64
select HAVE_STATIC_CALL
+ select HAVE_STATIC_CALL_INLINE if HAVE_STACK_VALIDATION
select HAVE_RSEQ
select HAVE_SYSCALL_TRACEPOINTS
select HAVE_UNSTABLE_SCHED_CLOCK
@@ -226,6 +227,7 @@ config X86
select RTC_MC146818_LIB
select SPARSE_IRQ
select SRCU
+ select STACK_VALIDATION if HAVE_STACK_VALIDATION && (HAVE_STATIC_CALL_INLINE || RETPOLINE)
select SYSCTL_EXCEPTION_TRACE
select THREAD_INFO_IN_TASK
select USER_STACKTRACE_SUPPORT
@@ -446,7 +448,6 @@ config GOLDFISH
config RETPOLINE
bool "Avoid speculative indirect branches in kernel"
default y
- select STACK_VALIDATION if HAVE_STACK_VALIDATION
help
Compile kernel with the retpoline compiler options to guard against
kernel-to-user data leaks by avoiding speculative indirect
--- a/arch/x86/include/asm/static_call.h
+++ b/arch/x86/include/asm/static_call.h
@@ -5,12 +5,23 @@
#include <asm/text-patching.h>

/*
+ * For CONFIG_HAVE_STATIC_CALL_INLINE, this is a temporary trampoline which
+ * uses the current value of the key->func pointer to do an indirect jump to
+ * the function. This trampoline is only used during boot, before the call
+ * sites get patched by static_call_update(). The name of this trampoline has
+ * a magical aspect: objtool uses it to find static call sites so it can create
+ * the .static_call_sites section.
+ *
* For CONFIG_HAVE_STATIC_CALL, this is a permanent trampoline which
* does a direct jump to the function. The direct jump gets patched by
* static_call_update().
+ *
+ * Having the trampoline in a special section forces GCC to emit a JMP.d32 when
+ * it does tail-call optimization on the call; since you cannot compute the
+ * relative displacement across sections.
*/
#define ARCH_DEFINE_STATIC_CALL_TRAMP(name, func) \
- asm(".pushsection .text, \"ax\" \n" \
+ asm(".pushsection .static_call.text, \"ax\" \n" \
".align 4 \n" \
".globl " STATIC_CALL_TRAMP_STR(name) " \n" \
STATIC_CALL_TRAMP_STR(name) ": \n" \
--- a/arch/x86/kernel/static_call.c
+++ b/arch/x86/kernel/static_call.c
@@ -6,7 +6,7 @@

static void __static_call_transform(void *insn, u8 opcode, void *func)
{
- const void *code = text_gen_insn(opcode, (long)insn, (long)func);
+ const void *code = text_gen_insn(opcode, insn, func);

if (WARN_ONCE(*(u8 *)insn != opcode,
"unexpected static call insn opcode 0x%x at %pS\n",
@@ -26,6 +26,9 @@ void arch_static_call_transform(void *si
if (tramp)
__static_call_transform(tramp, JMP32_INSN_OPCODE, func);

+ if (IS_ENABLED(CONFIG_HAVE_STATIC_CALL_INLINE) && site)
+ __static_call_transform(site, CALL_INSN_OPCODE, func);
+
mutex_unlock(&text_mutex);
}
EXPORT_SYMBOL_GPL(arch_static_call_transform);
--- a/arch/x86/kernel/vmlinux.lds.S
+++ b/arch/x86/kernel/vmlinux.lds.S
@@ -137,6 +137,7 @@ SECTIONS
IRQENTRY_TEXT
ALIGN_ENTRY_TEXT_END
SOFTIRQENTRY_TEXT
+ STATIC_CALL_TEXT
*(.fixup)
*(.gnu.warning)

--- a/include/asm-generic/vmlinux.lds.h
+++ b/include/asm-generic/vmlinux.lds.h
@@ -610,6 +610,12 @@
*(.softirqentry.text) \
__softirqentry_text_end = .;

+#define STATIC_CALL_TEXT \
+ ALIGN_FUNCTION(); \
+ __static_call_text_start = .; \
+ *(.static_call.text) \
+ __static_call_text_end = .;
+
/* Section used for early init (in .S files) */
#define HEAD_TEXT KEEP(*(.head.text))

--- /dev/null
+++ b/tools/include/linux/static_call_types.h
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _STATIC_CALL_TYPES_H
+#define _STATIC_CALL_TYPES_H
+
+#include <linux/types.h>
+#include <linux/stringify.h>
+
+#define STATIC_CALL_PREFIX __SC__
+#define STATIC_CALL_PREFIX_STR __stringify(STATIC_CALL_PREFIX)
+#define STATIC_CALL_PREFIX_LEN (sizeof(STATIC_CALL_PREFIX_STR) - 1)
+#define STATIC_CALL_KEY(name) __PASTE(STATIC_CALL_PREFIX, name)
+
+#define STATIC_CALL_TRAMP_PREFIX __SCT__
+#define STATIC_CALL_TRAMP_PREFIX_STR __stringify(STATIC_CALL_TRAMP_PREFIX)
+#define STATIC_CALL_TRAMP_PREFIX_LEN (sizeof(STATIC_CALL_TRAMP_PREFIX_STR) - 1)
+#define STATIC_CALL_TRAMP(name) __PASTE(STATIC_CALL_TRAMP_PREFIX, name)
+#define STATIC_CALL_TRAMP_STR(name) __stringify(STATIC_CALL_TRAMP(name))
+
+/*
+ * The static call site table needs to be created by external tooling (objtool
+ * or a compiler plugin).
+ */
+struct static_call_site {
+ s32 addr;
+ s32 key;
+};
+
+#endif /* _STATIC_CALL_TYPES_H */
--- a/tools/objtool/check.c
+++ b/tools/objtool/check.c
@@ -15,6 +15,7 @@

#include <linux/hashtable.h>
#include <linux/kernel.h>
+#include <linux/static_call_types.h>

#define FAKE_JUMP_OFFSET -1

@@ -423,6 +424,99 @@ static int add_dead_ends(struct objtool_
return 0;
}

+static int create_static_call_sections(struct objtool_file *file)
+{
+ struct section *sec, *rela_sec;
+ struct rela *rela;
+ struct static_call_site *site;
+ struct instruction *insn;
+ struct symbol *key_sym;
+ char *key_name, *tmp;
+ int idx;
+
+ sec = find_section_by_name(file->elf, ".static_call_sites");
+ if (sec) {
+ INIT_LIST_HEAD(&file->static_call_list);
+ WARN("file already has .static_call_sites section, skipping");
+ return 0;
+ }
+
+ if (list_empty(&file->static_call_list))
+ return 0;
+
+ idx = 0;
+ list_for_each_entry(insn, &file->static_call_list, static_call_node)
+ idx++;
+
+ sec = elf_create_section(file->elf, ".static_call_sites", SHF_WRITE,
+ sizeof(struct static_call_site), idx);
+ if (!sec)
+ return -1;
+
+ rela_sec = elf_create_rela_section(file->elf, sec);
+ if (!rela_sec)
+ return -1;
+
+ idx = 0;
+ list_for_each_entry(insn, &file->static_call_list, static_call_node) {
+
+ site = (struct static_call_site *)sec->data->d_buf + idx;
+ memset(site, 0, sizeof(struct static_call_site));
+
+ /* populate rela for 'addr' */
+ rela = malloc(sizeof(*rela));
+ if (!rela) {
+ perror("malloc");
+ return -1;
+ }
+ memset(rela, 0, sizeof(*rela));
+ rela->sym = insn->sec->sym;
+ rela->addend = insn->offset;
+ rela->type = R_X86_64_PC32;
+ rela->offset = idx * sizeof(struct static_call_site);
+ rela->sec = rela_sec;
+ elf_add_rela(file->elf, rela);
+
+ /* find key symbol */
+ key_name = strdup(insn->call_dest->name);
+ if (strncmp(key_name, STATIC_CALL_TRAMP_PREFIX_STR,
+ strlen(STATIC_CALL_TRAMP_PREFIX_STR))) {
+ WARN("static_call: trampoline name malformed: %s", key_name);
+ return -1;
+ }
+ tmp = key_name + STATIC_CALL_TRAMP_PREFIX_LEN - STATIC_CALL_PREFIX_LEN;
+ memcpy(tmp, STATIC_CALL_PREFIX_STR, STATIC_CALL_PREFIX_LEN);
+
+ key_sym = find_symbol_by_name(file->elf, tmp);
+ if (!key_sym) {
+ WARN("static_call: can't find static_call_key symbol: %s", tmp);
+ return -1;
+ }
+ free(key_name);
+
+ /* populate rela for 'key' */
+ rela = malloc(sizeof(*rela));
+ if (!rela) {
+ perror("malloc");
+ return -1;
+ }
+ memset(rela, 0, sizeof(*rela));
+ rela->sym = key_sym;
+ rela->addend = 0;
+ rela->type = R_X86_64_PC32;
+ rela->offset = idx * sizeof(struct static_call_site) + 4;
+ rela->sec = rela_sec;
+ elf_add_rela(file->elf, rela);
+
+ idx++;
+ }
+
+ if (elf_rebuild_rela_section(rela_sec))
+ return -1;
+
+ return 0;
+}
+
/*
* Warnings shouldn't be reported for ignored functions.
*/
@@ -1471,6 +1565,23 @@ static int read_intra_function_calls(str
return 0;
}

+static int read_static_call_tramps(struct objtool_file *file)
+{
+ struct section *sec;
+ struct symbol *func;
+
+ for_each_sec(file, sec) {
+ list_for_each_entry(func, &sec->symbol_list, list) {
+ if (func->bind == STB_GLOBAL &&
+ !strncmp(func->name, STATIC_CALL_TRAMP_PREFIX_STR,
+ strlen(STATIC_CALL_TRAMP_PREFIX_STR)))
+ func->static_call_tramp = true;
+ }
+ }
+
+ return 0;
+}
+
static void mark_rodata(struct objtool_file *file)
{
struct section *sec;
@@ -1550,6 +1661,10 @@ static int decode_sections(struct objtoo
if (ret)
return ret;

+ ret = read_static_call_tramps(file);
+ if (ret)
+ return ret;
+
return 0;
}

@@ -2354,6 +2469,11 @@ static int validate_branch(struct objtoo
if (dead_end_function(file, insn->call_dest))
return 0;

+ if (insn->type == INSN_CALL && insn->call_dest->static_call_tramp) {
+ list_add_tail(&insn->static_call_node,
+ &file->static_call_list);
+ }
+
break;

case INSN_JUMP_CONDITIONAL:
@@ -2707,12 +2827,13 @@ int check(const char *_objname, bool orc

objname = _objname;

- file.elf = elf_open_read(objname, orc ? O_RDWR : O_RDONLY);
+ file.elf = elf_open_read(objname, O_RDWR);
if (!file.elf)
return 1;

INIT_LIST_HEAD(&file.insn_list);
hash_init(file.insn_hash);
+ INIT_LIST_HEAD(&file.static_call_list);
file.c_file = find_section_by_name(file.elf, ".comment");
file.ignore_unreachables = no_unreachable;
file.hints = false;
@@ -2760,6 +2881,11 @@ int check(const char *_objname, bool orc
warnings += ret;
}

+ ret = create_static_call_sections(&file);
+ if (ret < 0)
+ goto out;
+ warnings += ret;
+
if (orc) {
ret = create_orc(&file);
if (ret < 0)
@@ -2768,7 +2894,9 @@ int check(const char *_objname, bool orc
ret = create_orc_sections(&file);
if (ret < 0)
goto out;
+ }

+ if (orc || !list_empty(&file.static_call_list)) {
ret = elf_write(file.elf);
if (ret < 0)
goto out;
--- a/tools/objtool/check.h
+++ b/tools/objtool/check.h
@@ -25,6 +25,7 @@ struct insn_state {
struct instruction {
struct list_head list;
struct hlist_node hash;
+ struct list_head static_call_node;
struct section *sec;
unsigned long offset;
unsigned int len;
@@ -52,6 +53,7 @@ struct objtool_file {
struct elf *elf;
struct list_head insn_list;
DECLARE_HASHTABLE(insn_hash, 20);
+ struct list_head static_call_list;
bool ignore_unreachables, c_file, hints, rodata;
};

--- a/tools/objtool/elf.c
+++ b/tools/objtool/elf.c
@@ -606,7 +606,7 @@ struct elf *elf_open_read(const char *na
}

struct section *elf_create_section(struct elf *elf, const char *name,
- size_t entsize, int nr)
+ unsigned int sh_flags, size_t entsize, int nr)
{
struct section *sec, *shstrtab;
size_t size = entsize * nr;
@@ -666,7 +666,7 @@ struct section *elf_create_section(struc
sec->sh.sh_entsize = entsize;
sec->sh.sh_type = SHT_PROGBITS;
sec->sh.sh_addralign = 1;
- sec->sh.sh_flags = SHF_ALLOC;
+ sec->sh.sh_flags = SHF_ALLOC | sh_flags;


/* Add section name to .shstrtab (or .strtab for Clang) */
@@ -719,7 +719,7 @@ struct section *elf_create_rela_section(
strcpy(relaname, ".rela");
strcat(relaname, base->name);

- sec = elf_create_section(elf, relaname, sizeof(GElf_Rela), 0);
+ sec = elf_create_section(elf, relaname, 0, sizeof(GElf_Rela), 0);
free(relaname);
if (!sec)
return NULL;
--- a/tools/objtool/elf.h
+++ b/tools/objtool/elf.h
@@ -56,6 +56,7 @@ struct symbol {
unsigned int len;
struct symbol *pfunc, *cfunc, *alias;
bool uaccess_safe;
+ bool static_call_tramp;
};

struct rela {
@@ -114,7 +115,7 @@ static inline u32 rela_hash(struct rela
}

struct elf *elf_open_read(const char *name, int flags);
-struct section *elf_create_section(struct elf *elf, const char *name, size_t entsize, int nr);
+struct section *elf_create_section(struct elf *elf, const char *name, unsigned int sh_flags, size_t entsize, int nr);
struct section *elf_create_rela_section(struct elf *elf, struct section *base);
void elf_add_rela(struct elf *elf, struct rela *rela);
int elf_write(const struct elf *elf);
--- a/tools/objtool/orc_gen.c
+++ b/tools/objtool/orc_gen.c
@@ -178,7 +178,7 @@ int create_orc_sections(struct objtool_f


/* create .orc_unwind_ip and .rela.orc_unwind_ip sections */
- sec = elf_create_section(file->elf, ".orc_unwind_ip", sizeof(int), idx);
+ sec = elf_create_section(file->elf, ".orc_unwind_ip", 0, sizeof(int), idx);
if (!sec)
return -1;

@@ -187,7 +187,7 @@ int create_orc_sections(struct objtool_f
return -1;

/* create .orc_unwind section */
- u_sec = elf_create_section(file->elf, ".orc_unwind",
+ u_sec = elf_create_section(file->elf, ".orc_unwind", 0,
sizeof(struct orc_entry), idx);

/* populate sections */
--- a/tools/objtool/sync-check.sh
+++ b/tools/objtool/sync-check.sh
@@ -7,6 +7,7 @@ arch/x86/include/asm/orc_types.h
arch/x86/include/asm/emulate_prefix.h
arch/x86/lib/x86-opcode-map.txt
arch/x86/tools/gen-insn-attr-x86.awk
+include/linux/static_call_types.h
'

check_2 () {