Merge branch 'x86-asm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86/asm changes from Ingo Molnar:
 "The one change that stands out is the alternatives patching change
  that prevents us from ever patching back instructions from SMP to UP:
  this simplifies things and speeds up CPU hotplug.

  Other than that it's smaller fixes, cleanups and improvements."

* 'x86-asm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86: Unspaghettize do_trap()
  x86_64: Work around old GAS bug
  x86: Use REP BSF unconditionally
  x86: Prefer TZCNT over BFS
  x86/64: Adjust types of temporaries used by ffs()/fls()/fls64()
  x86: Drop unnecessary kernel_eflags variable on 64-bit
  x86/smp: Don't ever patch back to UP if we unplug cpus
This commit is contained in:
Linus Torvalds 2012-10-01 10:46:27 -07:00
commit da8347969f
12 changed files with 104 additions and 202 deletions

View file

@ -2649,9 +2649,6 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
smart2= [HW]
Format: <io1>[,<io2>[,...,<io8>]]
smp-alt-once [X86-32,SMP] On a hotplug CPU system, only
attempt to substitute SMP alternatives once at boot.
smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
smsc-ircc2.ircc_sir= [HW] SIR base I/O port

View file

@ -60,7 +60,7 @@ extern void alternatives_smp_module_add(struct module *mod, char *name,
void *locks, void *locks_end,
void *text, void *text_end);
extern void alternatives_smp_module_del(struct module *mod);
extern void alternatives_smp_switch(int smp);
extern void alternatives_enable_smp(void);
extern int alternatives_text_reserved(void *start, void *end);
extern bool skip_smp_alternatives;
#else
@ -68,7 +68,7 @@ static inline void alternatives_smp_module_add(struct module *mod, char *name,
void *locks, void *locks_end,
void *text, void *text_end) {}
static inline void alternatives_smp_module_del(struct module *mod) {}
static inline void alternatives_smp_switch(int smp) {}
static inline void alternatives_enable_smp(void) {}
static inline int alternatives_text_reserved(void *start, void *end)
{
return 0;

View file

@ -355,7 +355,7 @@ static int test_bit(int nr, const volatile unsigned long *addr);
*/
static inline unsigned long __ffs(unsigned long word)
{
asm("bsf %1,%0"
asm("rep; bsf %1,%0"
: "=r" (word)
: "rm" (word));
return word;
@ -369,7 +369,7 @@ static inline unsigned long __ffs(unsigned long word)
*/
static inline unsigned long ffz(unsigned long word)
{
asm("bsf %1,%0"
asm("rep; bsf %1,%0"
: "=r" (word)
: "r" (~word));
return word;
@ -417,10 +417,9 @@ static inline int ffs(int x)
* We cannot do this on 32 bits because at the very least some
* 486 CPUs did not behave this way.
*/
long tmp = -1;
asm("bsfl %1,%0"
: "=r" (r)
: "rm" (x), "0" (tmp));
: "rm" (x), "0" (-1));
#elif defined(CONFIG_X86_CMOV)
asm("bsfl %1,%0\n\t"
"cmovzl %2,%0"
@ -459,10 +458,9 @@ static inline int fls(int x)
* We cannot do this on 32 bits because at the very least some
* 486 CPUs did not behave this way.
*/
long tmp = -1;
asm("bsrl %1,%0"
: "=r" (r)
: "rm" (x), "0" (tmp));
: "rm" (x), "0" (-1));
#elif defined(CONFIG_X86_CMOV)
asm("bsrl %1,%0\n\t"
"cmovzl %2,%0"
@ -490,13 +488,13 @@ static inline int fls(int x)
#ifdef CONFIG_X86_64
static __always_inline int fls64(__u64 x)
{
long bitpos = -1;
int bitpos = -1;
/*
* AMD64 says BSRQ won't clobber the dest reg if x==0; Intel64 says the
* dest reg is undefined if x==0, but their CPU architect says its
* value is written to set it to the same as before.
*/
asm("bsrq %1,%0"
asm("bsrq %1,%q0"
: "+r" (bitpos)
: "rm" (x));
return bitpos + 1;

View file

@ -49,38 +49,36 @@ For 32-bit we have the following conventions - kernel is built with
#include "dwarf2.h"
/*
* 64-bit system call stack frame layout defines and helpers, for
* assembly code (note that the seemingly unnecessary parentheses
* are to prevent cpp from inserting spaces in expressions that get
* passed to macros):
* 64-bit system call stack frame layout defines and helpers,
* for assembly code:
*/
#define R15 (0)
#define R14 (8)
#define R13 (16)
#define R12 (24)
#define RBP (32)
#define RBX (40)
#define R15 0
#define R14 8
#define R13 16
#define R12 24
#define RBP 32
#define RBX 40
/* arguments: interrupts/non tracing syscalls only save up to here: */
#define R11 (48)
#define R10 (56)
#define R9 (64)
#define R8 (72)
#define RAX (80)
#define RCX (88)
#define RDX (96)
#define RSI (104)
#define RDI (112)
#define ORIG_RAX (120) /* + error_code */
#define R11 48
#define R10 56
#define R9 64
#define R8 72
#define RAX 80
#define RCX 88
#define RDX 96
#define RSI 104
#define RDI 112
#define ORIG_RAX 120 /* + error_code */
/* end of arguments */
/* cpu exception frame or undefined in case of fast syscall: */
#define RIP (128)
#define CS (136)
#define EFLAGS (144)
#define RSP (152)
#define SS (160)
#define RIP 128
#define CS 136
#define EFLAGS 144
#define RSP 152
#define SS 160
#define ARGOFFSET R11
#define SWFRAME ORIG_RAX

View file

@ -423,7 +423,6 @@ DECLARE_INIT_PER_CPU(irq_stack_union);
DECLARE_PER_CPU(char *, irq_stack_ptr);
DECLARE_PER_CPU(unsigned int, irq_count);
extern unsigned long kernel_eflags;
extern asmlinkage void ignore_sysret(void);
#else /* X86_64 */
#ifdef CONFIG_CC_STACKPROTECTOR

View file

@ -23,19 +23,6 @@
#define MAX_PATCH_LEN (255-1)
#ifdef CONFIG_HOTPLUG_CPU
static int smp_alt_once;
static int __init bootonly(char *str)
{
smp_alt_once = 1;
return 1;
}
__setup("smp-alt-boot", bootonly);
#else
#define smp_alt_once 1
#endif
static int __initdata_or_module debug_alternative;
static int __init debug_alt(char *str)
@ -326,9 +313,6 @@ static void alternatives_smp_unlock(const s32 *start, const s32 *end,
{
const s32 *poff;
if (noreplace_smp)
return;
mutex_lock(&text_mutex);
for (poff = start; poff < end; poff++) {
u8 *ptr = (u8 *)poff + *poff;
@ -359,7 +343,7 @@ struct smp_alt_module {
};
static LIST_HEAD(smp_alt_modules);
static DEFINE_MUTEX(smp_alt);
static int smp_mode = 1; /* protected by smp_alt */
static bool uniproc_patched = false; /* protected by smp_alt */
void __init_or_module alternatives_smp_module_add(struct module *mod,
char *name,
@ -368,19 +352,18 @@ void __init_or_module alternatives_smp_module_add(struct module *mod,
{
struct smp_alt_module *smp;
if (noreplace_smp)
return;
mutex_lock(&smp_alt);
if (!uniproc_patched)
goto unlock;
if (smp_alt_once) {
if (boot_cpu_has(X86_FEATURE_UP))
alternatives_smp_unlock(locks, locks_end,
text, text_end);
return;
}
if (num_possible_cpus() == 1)
/* Don't bother remembering, we'll never have to undo it. */
goto smp_unlock;
smp = kzalloc(sizeof(*smp), GFP_KERNEL);
if (NULL == smp)
return; /* we'll run the (safe but slow) SMP code then ... */
/* we'll run the (safe but slow) SMP code then ... */
goto unlock;
smp->mod = mod;
smp->name = name;
@ -392,11 +375,10 @@ void __init_or_module alternatives_smp_module_add(struct module *mod,
__func__, smp->locks, smp->locks_end,
smp->text, smp->text_end, smp->name);
mutex_lock(&smp_alt);
list_add_tail(&smp->next, &smp_alt_modules);
if (boot_cpu_has(X86_FEATURE_UP))
alternatives_smp_unlock(smp->locks, smp->locks_end,
smp->text, smp->text_end);
smp_unlock:
alternatives_smp_unlock(locks, locks_end, text, text_end);
unlock:
mutex_unlock(&smp_alt);
}
@ -404,24 +386,18 @@ void __init_or_module alternatives_smp_module_del(struct module *mod)
{
struct smp_alt_module *item;
if (smp_alt_once || noreplace_smp)
return;
mutex_lock(&smp_alt);
list_for_each_entry(item, &smp_alt_modules, next) {
if (mod != item->mod)
continue;
list_del(&item->next);
mutex_unlock(&smp_alt);
DPRINTK("%s: %s\n", __func__, item->name);
kfree(item);
return;
break;
}
mutex_unlock(&smp_alt);
}
bool skip_smp_alternatives;
void alternatives_smp_switch(int smp)
void alternatives_enable_smp(void)
{
struct smp_alt_module *mod;
@ -436,34 +412,21 @@ void alternatives_smp_switch(int smp)
pr_info("lockdep: fixing up alternatives\n");
#endif
if (noreplace_smp || smp_alt_once || skip_smp_alternatives)
return;
BUG_ON(!smp && (num_online_cpus() > 1));
/* Why bother if there are no other CPUs? */
BUG_ON(num_possible_cpus() == 1);
mutex_lock(&smp_alt);
/*
* Avoid unnecessary switches because it forces JIT based VMs to
* throw away all cached translations, which can be quite costly.
*/
if (smp == smp_mode) {
/* nothing */
} else if (smp) {
if (uniproc_patched) {
pr_info("switching to SMP code\n");
BUG_ON(num_online_cpus() != 1);
clear_cpu_cap(&boot_cpu_data, X86_FEATURE_UP);
clear_cpu_cap(&cpu_data(0), X86_FEATURE_UP);
list_for_each_entry(mod, &smp_alt_modules, next)
alternatives_smp_lock(mod->locks, mod->locks_end,
mod->text, mod->text_end);
} else {
pr_info("switching to UP code\n");
set_cpu_cap(&boot_cpu_data, X86_FEATURE_UP);
set_cpu_cap(&cpu_data(0), X86_FEATURE_UP);
list_for_each_entry(mod, &smp_alt_modules, next)
alternatives_smp_unlock(mod->locks, mod->locks_end,
mod->text, mod->text_end);
uniproc_patched = false;
}
smp_mode = smp;
mutex_unlock(&smp_alt);
}
@ -540,40 +503,22 @@ void __init alternative_instructions(void)
apply_alternatives(__alt_instructions, __alt_instructions_end);
/* switch to patch-once-at-boottime-only mode and free the
* tables in case we know the number of CPUs will never ever
* change */
#ifdef CONFIG_HOTPLUG_CPU
if (num_possible_cpus() < 2)
smp_alt_once = 1;
#endif
#ifdef CONFIG_SMP
if (smp_alt_once) {
if (1 == num_possible_cpus()) {
pr_info("switching to UP code\n");
set_cpu_cap(&boot_cpu_data, X86_FEATURE_UP);
set_cpu_cap(&cpu_data(0), X86_FEATURE_UP);
alternatives_smp_unlock(__smp_locks, __smp_locks_end,
_text, _etext);
}
} else {
/* Patch to UP if other cpus not imminent. */
if (!noreplace_smp && (num_present_cpus() == 1 || setup_max_cpus <= 1)) {
uniproc_patched = true;
alternatives_smp_module_add(NULL, "core kernel",
__smp_locks, __smp_locks_end,
_text, _etext);
/* Only switch to UP mode if we don't immediately boot others */
if (num_present_cpus() == 1 || setup_max_cpus <= 1)
alternatives_smp_switch(0);
}
#endif
apply_paravirt(__parainstructions, __parainstructions_end);
if (smp_alt_once)
if (!uniproc_patched || num_possible_cpus() == 1)
free_init_pages("SMP alternatives",
(unsigned long)__smp_locks,
(unsigned long)__smp_locks_end);
#endif
apply_paravirt(__parainstructions, __parainstructions_end);
restart_nmi();
}

View file

@ -1116,8 +1116,6 @@ void syscall_init(void)
X86_EFLAGS_TF|X86_EFLAGS_DF|X86_EFLAGS_IF|X86_EFLAGS_IOPL);
}
unsigned long kernel_eflags;
/*
* Copies of the original ist values from the tss are only accessed during
* debugging, no special alignment required.
@ -1299,8 +1297,6 @@ void __cpuinit cpu_init(void)
fpu_init();
xsave_init();
raw_local_save_flags(kernel_eflags);
if (is_uv_system())
uv_cpu_init();
}

View file

@ -445,15 +445,15 @@ ENDPROC(native_usergs_sysret64)
.macro SAVE_ARGS_IRQ
cld
/* start from rbp in pt_regs and jump over */
movq_cfi rdi, RDI-RBP
movq_cfi rsi, RSI-RBP
movq_cfi rdx, RDX-RBP
movq_cfi rcx, RCX-RBP
movq_cfi rax, RAX-RBP
movq_cfi r8, R8-RBP
movq_cfi r9, R9-RBP
movq_cfi r10, R10-RBP
movq_cfi r11, R11-RBP
movq_cfi rdi, (RDI-RBP)
movq_cfi rsi, (RSI-RBP)
movq_cfi rdx, (RDX-RBP)
movq_cfi rcx, (RCX-RBP)
movq_cfi rax, (RAX-RBP)
movq_cfi r8, (R8-RBP)
movq_cfi r9, (R9-RBP)
movq_cfi r10, (R10-RBP)
movq_cfi r11, (R11-RBP)
/* Save rbp so that we can unwind from get_irq_regs() */
movq_cfi rbp, 0
@ -487,7 +487,7 @@ ENDPROC(native_usergs_sysret64)
.endm
ENTRY(save_rest)
PARTIAL_FRAME 1 REST_SKIP+8
PARTIAL_FRAME 1 (REST_SKIP+8)
movq 5*8+16(%rsp), %r11 /* save return address */
movq_cfi rbx, RBX+16
movq_cfi rbp, RBP+16
@ -543,7 +543,7 @@ ENTRY(ret_from_fork)
LOCK ; btr $TIF_FORK,TI_flags(%r8)
pushq_cfi kernel_eflags(%rip)
pushq_cfi $0x0002
popfq_cfi # reset kernel eflags
call schedule_tail # rdi: 'prev' task parameter

View file

@ -665,7 +665,8 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu, struct task_struct *idle)
unsigned long boot_error = 0;
int timeout;
alternatives_smp_switch(1);
/* Just in case we booted with a single CPU. */
alternatives_enable_smp();
idle->thread.sp = (unsigned long) (((struct pt_regs *)
(THREAD_SIZE + task_stack_page(idle))) - 1);
@ -1053,20 +1054,6 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus)
preempt_enable();
}
void arch_disable_nonboot_cpus_begin(void)
{
/*
* Avoid the smp alternatives switch during the disable_nonboot_cpus().
* In the suspend path, we will be back in the SMP mode shortly anyways.
*/
skip_smp_alternatives = true;
}
void arch_disable_nonboot_cpus_end(void)
{
skip_smp_alternatives = false;
}
void arch_enable_nonboot_cpus_begin(void)
{
set_mtrr_aps_delayed_init();
@ -1256,9 +1243,6 @@ void native_cpu_die(unsigned int cpu)
if (per_cpu(cpu_state, cpu) == CPU_DEAD) {
if (system_state == SYSTEM_RUNNING)
pr_info("CPU %u is now offline\n", cpu);
if (1 == num_online_cpus())
alternatives_smp_switch(0);
return;
}
msleep(100);

View file

@ -108,30 +108,45 @@ static inline void preempt_conditional_cli(struct pt_regs *regs)
dec_preempt_count();
}
static int __kprobes
do_trap_no_signal(struct task_struct *tsk, int trapnr, char *str,
struct pt_regs *regs, long error_code)
{
#ifdef CONFIG_X86_32
if (regs->flags & X86_VM_MASK) {
/*
* Traps 0, 1, 3, 4, and 5 should be forwarded to vm86.
* On nmi (interrupt 2), do_trap should not be called.
*/
if (trapnr < X86_TRAP_UD) {
if (!handle_vm86_trap((struct kernel_vm86_regs *) regs,
error_code, trapnr))
return 0;
}
return -1;
}
#endif
if (!user_mode(regs)) {
if (!fixup_exception(regs)) {
tsk->thread.error_code = error_code;
tsk->thread.trap_nr = trapnr;
die(str, regs, error_code);
}
return 0;
}
return -1;
}
static void __kprobes
do_trap(int trapnr, int signr, char *str, struct pt_regs *regs,
long error_code, siginfo_t *info)
{
struct task_struct *tsk = current;
#ifdef CONFIG_X86_32
if (regs->flags & X86_VM_MASK) {
/*
* traps 0, 1, 3, 4, and 5 should be forwarded to vm86.
* On nmi (interrupt 2), do_trap should not be called.
*/
if (trapnr < X86_TRAP_UD)
goto vm86_trap;
goto trap_signal;
}
#endif
if (!user_mode(regs))
goto kernel_trap;
#ifdef CONFIG_X86_32
trap_signal:
#endif
if (!do_trap_no_signal(tsk, trapnr, str, regs, error_code))
return;
/*
* We want error_code and trap_nr set for userspace faults and
* kernelspace faults which result in die(), but not
@ -159,23 +174,6 @@ do_trap(int trapnr, int signr, char *str, struct pt_regs *regs,
force_sig_info(signr, info, tsk);
else
force_sig(signr, tsk);
return;
kernel_trap:
if (!fixup_exception(regs)) {
tsk->thread.error_code = error_code;
tsk->thread.trap_nr = trapnr;
die(str, regs, error_code);
}
return;
#ifdef CONFIG_X86_32
vm86_trap:
if (handle_vm86_trap((struct kernel_vm86_regs *) regs,
error_code, trapnr))
goto trap_signal;
return;
#endif
}
#define DO_ERROR(trapnr, signr, str, name) \

View file

@ -377,7 +377,8 @@ static int __cpuinit xen_cpu_up(unsigned int cpu, struct task_struct *idle)
return rc;
if (num_online_cpus() == 1)
alternatives_smp_switch(1);
/* Just in case we booted with a single CPU. */
alternatives_enable_smp();
rc = xen_smp_intr_init(cpu);
if (rc)
@ -424,9 +425,6 @@ static void xen_cpu_die(unsigned int cpu)
unbind_from_irqhandler(per_cpu(xen_irq_work, cpu), NULL);
xen_uninit_lock_cpu(cpu);
xen_teardown_timer(cpu);
if (num_online_cpus() == 1)
alternatives_smp_switch(0);
}
static void __cpuinit xen_play_dead(void) /* used only with HOTPLUG_CPU */

View file

@ -447,14 +447,6 @@ EXPORT_SYMBOL_GPL(cpu_up);
#ifdef CONFIG_PM_SLEEP_SMP
static cpumask_var_t frozen_cpus;
void __weak arch_disable_nonboot_cpus_begin(void)
{
}
void __weak arch_disable_nonboot_cpus_end(void)
{
}
int disable_nonboot_cpus(void)
{
int cpu, first_cpu, error = 0;
@ -466,7 +458,6 @@ int disable_nonboot_cpus(void)
* with the userspace trying to use the CPU hotplug at the same time
*/
cpumask_clear(frozen_cpus);
arch_disable_nonboot_cpus_begin();
printk("Disabling non-boot CPUs ...\n");
for_each_online_cpu(cpu) {
@ -482,8 +473,6 @@ int disable_nonboot_cpus(void)
}
}
arch_disable_nonboot_cpus_end();
if (!error) {
BUG_ON(num_online_cpus() > 1);
/* Make sure the CPUs won't be enabled by someone else */