x86: x86 i387 cleanup

/*

 * Copyright 2002 Andi Kleen, SuSE Labs.

 * FXSAVE<->i387 conversion support. Based on code by Gareth Hughes.

 * This is used for ptrace, signals and coredumps in 32bit emulation.

 */

#include <linux/sched.h>

#include <asm/sigcontext32.h>

#include <asm/processor.h>

#include <asm/uaccess.h>

#include <asm/i387.h>

static inline unsigned short twd_i387_to_fxsr(unsigned short twd)

{

	unsigned int tmp; /* to avoid 16 bit prefixes in the code */

	/* Transform each pair of bits into 01 (valid) or 00 (empty) */

	tmp = ~twd;

	tmp = (tmp | (tmp>>1)) & 0x5555; /* 0V0V0V0V0V0V0V0V */

	/* and move the valid bits to the lower byte. */

	tmp = (tmp | (tmp >> 1)) & 0x3333; /* 00VV00VV00VV00VV */

	tmp = (tmp | (tmp >> 2)) & 0x0f0f; /* 0000VVVV0000VVVV */

	tmp = (tmp | (tmp >> 4)) & 0x00ff; /* 00000000VVVVVVVV */

	return tmp;

}

#define FPREG_ADDR(f, n)	((void *)&(f)->st_space + (n) * 16);

#define FP_EXP_TAG_VALID	0

#define FP_EXP_TAG_ZERO		1

#define FP_EXP_TAG_SPECIAL	2

#define FP_EXP_TAG_EMPTY	3

static inline unsigned long twd_fxsr_to_i387(struct i387_fxsave_struct *fxsave)

{

	struct _fpxreg *st;

	unsigned long tos = (fxsave->swd >> 11) & 7;

	unsigned long twd = (unsigned long) fxsave->twd;

	unsigned long tag;

	unsigned long ret = 0xffff0000;

	int i;

	for (i = 0; i < 8; i++, twd >>= 1) {

		if (twd & 0x1) {

			st = FPREG_ADDR(fxsave, (i - tos) & 7);

			switch (st->exponent & 0x7fff) {

			case 0x7fff:

				tag = FP_EXP_TAG_SPECIAL;

				break;

			case 0x0000:

				if (!st->significand[0] &&

				    !st->significand[1] &&

				    !st->significand[2] &&

				    !st->significand[3])

					tag = FP_EXP_TAG_ZERO;

				else

					tag = FP_EXP_TAG_SPECIAL;

				break;

			default:

				if (st->significand[3] & 0x8000)

					tag = FP_EXP_TAG_VALID;

				else

					tag = FP_EXP_TAG_SPECIAL;

				break;

			}

		} else {

			tag = FP_EXP_TAG_EMPTY;

		}

		ret |= tag << (2 * i);

	}

	return ret;

}

#define G(num, val) err |= __get_user(val, num + (u32 __user *)buf)

static inline int convert_fxsr_from_user(struct i387_fxsave_struct *fxsave,

					 struct _fpstate_ia32 __user *buf)

{

	struct _fpxreg *to;

	struct _fpreg __user *from;

	int i, err = 0;

	u32 v;

	G(0, fxsave->cwd);

	G(1, fxsave->swd);

	G(2, fxsave->twd);

	fxsave->twd = twd_i387_to_fxsr(fxsave->twd);

	G(3, fxsave->rip);

	G(4, v);

	/* cs ignored */

	fxsave->fop = v>>16;

	G(5, fxsave->rdp);

	/* 6: ds ignored */

	if (err)

		return -1;

	to = (struct _fpxreg *)&fxsave->st_space[0];

	from = &buf->_st[0];

	for (i = 0; i < 8; i++, to++, from++) {

		if (__copy_from_user(to, from, sizeof(*from)))

			return -1;

	}

	return 0;

}

#define P(num, val) err |= __put_user(val, num + (u32 __user *)buf)

static inline int convert_fxsr_to_user(struct _fpstate_ia32 __user *buf,

				       struct i387_fxsave_struct *fxsave,

				       struct pt_regs *regs,

				       struct task_struct *tsk)

{

	struct _fpreg __user *to;

	struct _fpxreg *from;

	int i, err = 0;

	u16 cs, ds;

	if (tsk == current) {

		/*

		 * should be actually ds/cs at fpu exception time, but

		 * that information is not available in 64bit mode.

		 */

		asm("movw %%ds,%0 " : "=r" (ds));

		asm("movw %%cs,%0 " : "=r" (cs));

	} else {

		 /* ptrace. task has stopped. */

		ds = tsk->thread.ds;

		cs = regs->cs;

	}

	P(0, (u32)fxsave->cwd | 0xffff0000);

	P(1, (u32)fxsave->swd | 0xffff0000);

	P(2, twd_fxsr_to_i387(fxsave));

	P(3, (u32)fxsave->rip);

	P(4,  cs | ((u32)fxsave->fop) << 16);

	P(5, fxsave->rdp);

	P(6, 0xffff0000 | ds);

	if (err)

		return -1;

	to = &buf->_st[0];

	from = (struct _fpxreg *) &fxsave->st_space[0];

	for (i = 0; i < 8; i++, to++, from++) {

		if (__copy_to_user(to, from, sizeof(*to)))

			return -1;

	}

	return 0;

}

int restore_i387_ia32(struct task_struct *tsk,

		      struct _fpstate_ia32 __user *buf, int fsave)

{

	clear_fpu(tsk);

	if (!fsave) {

		if (__copy_from_user(&tsk->thread.i387.fxsave,

				     &buf->_fxsr_env[0],

				     sizeof(struct i387_fxsave_struct)))

			return -1;

		tsk->thread.i387.fxsave.mxcsr &= mxcsr_feature_mask;

		set_stopped_child_used_math(tsk);

	}

	return convert_fxsr_from_user(&tsk->thread.i387.fxsave, buf);

}

int save_i387_ia32(struct task_struct *tsk, struct _fpstate_ia32 __user *buf,

		   struct pt_regs *regs, int fsave)

{

	int err = 0;

	init_fpu(tsk);

	if (convert_fxsr_to_user(buf, &tsk->thread.i387.fxsave, regs, tsk))

		return -1;

	if (fsave)

		return 0;

	err |= __put_user(tsk->thread.i387.fxsave.swd, &buf->status);

	err |= __put_user(X86_FXSR_MAGIC, &buf->magic);

	err |= __copy_to_user(&buf->_fxsr_env[0], &tsk->thread.i387.fxsave,

			      sizeof(struct i387_fxsave_struct));

	return err ? -1 : 1;

}

/*

 *  Copyright (C) 1994 Linus Torvalds

 *  Copyright (C) 2002 Andi Kleen, SuSE Labs

 *

 *  Pentium III FXSR, SSE support

 *  General FPU state handling cleanups

 *	Gareth Hughes <gareth@valinux.com>, May 2000

 * 

 *  x86-64 rework 2002 Andi Kleen. 

 *  Does direct fxsave in and out of user space now for signal handlers.

 *  All the FSAVE<->FXSAVE conversion code has been moved to the 32bit emulation,

 *  the 64bit user space sees a FXSAVE frame directly. 

 */

#include <linux/sched.h>

#include <linux/init.h>

#include <asm/processor.h>

#include <asm/i387.h>

#include <asm/sigcontext.h>

#include <asm/user.h>

#include <asm/ptrace.h>

#include <asm/uaccess.h>

unsigned int mxcsr_feature_mask __read_mostly = 0xffffffff;

void mxcsr_feature_mask_init(void)

{

	unsigned int mask;

	clts();

	memset(&current->thread.i387.fxsave, 0, sizeof(struct i387_fxsave_struct));

	asm volatile("fxsave %0" : : "m" (current->thread.i387.fxsave));

	mask = current->thread.i387.fxsave.mxcsr_mask;

	if (mask == 0) mask = 0x0000ffbf;

	mxcsr_feature_mask &= mask;

	stts();

}

/*

 * Called at bootup to set up the initial FPU state that is later cloned

 * into all processes.

 */

void __cpuinit fpu_init(void)

{

	unsigned long oldcr0 = read_cr0();

	extern void __bad_fxsave_alignment(void);

	if (offsetof(struct task_struct, thread.i387.fxsave) & 15)

		__bad_fxsave_alignment();

	set_in_cr4(X86_CR4_OSFXSR);

	set_in_cr4(X86_CR4_OSXMMEXCPT);

	write_cr0(oldcr0 & ~((1UL<<3)|(1UL<<2))); /* clear TS and EM */

	mxcsr_feature_mask_init();

	/* clean state in init */

	current_thread_info()->status = 0;

	clear_used_math();

}

void init_fpu(struct task_struct *child)

{

	if (tsk_used_math(child)) {

		if (child == current)

			unlazy_fpu(child);

		return;

	}	

	memset(&child->thread.i387.fxsave, 0, sizeof(struct i387_fxsave_struct));

	child->thread.i387.fxsave.cwd = 0x37f;

	child->thread.i387.fxsave.mxcsr = 0x1f80;

	/* only the device not available exception or ptrace can call init_fpu */

	set_stopped_child_used_math(child);

}

/*

 * ptrace request handlers.

 */

int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *tsk)

{

	init_fpu(tsk);

	return __copy_to_user(buf, &tsk->thread.i387.fxsave,

			       sizeof(struct user_i387_struct)) ? -EFAULT : 0;

}

int set_fpregs(struct task_struct *tsk, struct user_i387_struct __user *buf)

{

	if (__copy_from_user(&tsk->thread.i387.fxsave, buf, 

			     sizeof(struct user_i387_struct)))

		return -EFAULT;

		return 0;

}

/*

 * FPU state for core dumps.

 */

int dump_fpu( struct pt_regs *regs, struct user_i387_struct *fpu )

{

	struct task_struct *tsk = current;

	if (!used_math())

		return 0;

	unlazy_fpu(tsk);

	memcpy(fpu, &tsk->thread.i387.fxsave, sizeof(struct user_i387_struct)); 

	return 1; 

}

int dump_task_fpu(struct task_struct *tsk, struct user_i387_struct *fpu)

{

	int fpvalid = !!tsk_used_math(tsk);

	if (fpvalid) {

		if (tsk == current)

			unlazy_fpu(tsk);

		memcpy(fpu, &tsk->thread.i387.fxsave, sizeof(struct user_i387_struct)); 	

}

	return fpvalid;

}

#include "i387_64.h"

/*

 * Copyright (C) 1994 Linus Torvalds

 *

 * Pentium III FXSR, SSE support

 * General FPU state handling cleanups

 *	Gareth Hughes <gareth@valinux.com>, May 2000

 * x86-64 work by Andi Kleen 2002

 */

#ifndef _ASM_X86_I387_H

#define _ASM_X86_I387_H

#include <linux/sched.h>

#include <linux/kernel_stat.h>

#include <linux/regset.h>

#include <asm/processor.h>

#include <asm/sigcontext.h>

#include <asm/user.h>

#include <asm/uaccess.h>

extern void fpu_init(void);

extern unsigned int mxcsr_feature_mask;

extern void mxcsr_feature_mask_init(void);

extern void init_fpu(struct task_struct *child);

extern asmlinkage void math_state_restore(void);

extern user_regset_active_fn fpregs_active, xfpregs_active;

extern user_regset_get_fn fpregs_get, xfpregs_get, fpregs_soft_get;

extern user_regset_set_fn fpregs_set, xfpregs_set, fpregs_soft_set;

#ifdef CONFIG_IA32_EMULATION

struct _fpstate_ia32;

extern int save_i387_ia32(struct _fpstate_ia32 __user *buf);

extern int restore_i387_ia32(struct _fpstate_ia32 __user *buf);

#endif

#ifdef CONFIG_X86_64

/* Ignore delayed exceptions from user space */

static inline void tolerant_fwait(void)

{

	asm volatile("1: fwait\n"

		     "2:\n"

		     "   .section __ex_table,\"a\"\n"

		     "	.align 8\n"

		     "	.quad 1b,2b\n"

		     "	.previous\n");

}

static inline int restore_fpu_checking(struct i387_fxsave_struct *fx)

{

	int err;

	asm volatile("1:  rex64/fxrstor (%[fx])\n\t"

		     "2:\n"

		     ".section .fixup,\"ax\"\n"

		     "3:  movl $-1,%[err]\n"

		     "    jmp  2b\n"

		     ".previous\n"

		     ".section __ex_table,\"a\"\n"

		     "   .align 8\n"

		     "   .quad  1b,3b\n"

		     ".previous"

		     : [err] "=r" (err)

#if 0 /* See comment in __save_init_fpu() below. */

		     : [fx] "r" (fx), "m" (*fx), "0" (0));

#else

		     : [fx] "cdaSDb" (fx), "m" (*fx), "0" (0));

#endif

	if (unlikely(err))

		init_fpu(current);

	return err;

}

#define X87_FSW_ES (1 << 7)	/* Exception Summary */

/* AMD CPUs don't save/restore FDP/FIP/FOP unless an exception

   is pending. Clear the x87 state here by setting it to fixed

   values. The kernel data segment can be sometimes 0 and sometimes

   new user value. Both should be ok.

   Use the PDA as safe address because it should be already in L1. */

static inline void clear_fpu_state(struct i387_fxsave_struct *fx)

{

	if (unlikely(fx->swd & X87_FSW_ES))

		 asm volatile("fnclex");

	alternative_input(ASM_NOP8 ASM_NOP2,

		     "    emms\n"		/* clear stack tags */

		     "    fildl %%gs:0",	/* load to clear state */

		     X86_FEATURE_FXSAVE_LEAK);

}

static inline int save_i387_checking(struct i387_fxsave_struct __user *fx)

{

	int err;

	asm volatile("1:  rex64/fxsave (%[fx])\n\t"

		     "2:\n"

		     ".section .fixup,\"ax\"\n"

		     "3:  movl $-1,%[err]\n"

		     "    jmp  2b\n"

		     ".previous\n"

		     ".section __ex_table,\"a\"\n"

		     "   .align 8\n"

		     "   .quad  1b,3b\n"

		     ".previous"

		     : [err] "=r" (err), "=m" (*fx)

#if 0 /* See comment in __fxsave_clear() below. */

		     : [fx] "r" (fx), "0" (0));

#else

		     : [fx] "cdaSDb" (fx), "0" (0));

#endif

	if (unlikely(err) && __clear_user(fx, sizeof(struct i387_fxsave_struct)))

		err = -EFAULT;

	/* No need to clear here because the caller clears USED_MATH */

	return err;

}

static inline void __save_init_fpu(struct task_struct *tsk)

{

	/* Using "rex64; fxsave %0" is broken because, if the memory operand

	   uses any extended registers for addressing, a second REX prefix

	   will be generated (to the assembler, rex64 followed by semicolon

	   is a separate instruction), and hence the 64-bitness is lost. */

#if 0

	/* Using "fxsaveq %0" would be the ideal choice, but is only supported

	   starting with gas 2.16. */

	__asm__ __volatile__("fxsaveq %0"

			     : "=m" (tsk->thread.i387.fxsave));

#elif 0

	/* Using, as a workaround, the properly prefixed form below isn't

	   accepted by any binutils version so far released, complaining that

	   the same type of prefix is used twice if an extended register is

	   needed for addressing (fix submitted to mainline 2005-11-21). */

	__asm__ __volatile__("rex64/fxsave %0"

			     : "=m" (tsk->thread.i387.fxsave));

#else

	/* This, however, we can work around by forcing the compiler to select

	   an addressing mode that doesn't require extended registers. */

	__asm__ __volatile__("rex64/fxsave %P2(%1)"

			     : "=m" (tsk->thread.i387.fxsave)

			     : "cdaSDb" (tsk),

				"i" (offsetof(__typeof__(*tsk),

					      thread.i387.fxsave)));

#endif

	clear_fpu_state(&tsk->thread.i387.fxsave);

	task_thread_info(tsk)->status &= ~TS_USEDFPU;

}

/*

 * Signal frame handlers.

 */

static inline int save_i387(struct _fpstate __user *buf)

{

	struct task_struct *tsk = current;

	int err = 0;

	BUILD_BUG_ON(sizeof(struct user_i387_struct) !=

			sizeof(tsk->thread.i387.fxsave));

	if ((unsigned long)buf % 16)

		printk("save_i387: bad fpstate %p\n", buf);

	if (!used_math())

		return 0;

	clear_used_math(); /* trigger finit */

	if (task_thread_info(tsk)->status & TS_USEDFPU) {

		err = save_i387_checking((struct i387_fxsave_struct __user *)buf);

		if (err) return err;

		task_thread_info(tsk)->status &= ~TS_USEDFPU;

		stts();

	} else {

		if (__copy_to_user(buf, &tsk->thread.i387.fxsave,

				   sizeof(struct i387_fxsave_struct)))

			return -1;

	}

	return 1;

}

/*

 * This restores directly out of user space. Exceptions are handled.

 */

static inline int restore_i387(struct _fpstate __user *buf)

{

	set_used_math();

	if (!(task_thread_info(current)->status & TS_USEDFPU)) {

		clts();

		task_thread_info(current)->status |= TS_USEDFPU;

	}

	return restore_fpu_checking((__force struct i387_fxsave_struct *)buf);

}

#else  /* CONFIG_X86_32 */

static inline void tolerant_fwait(void)

{

	asm volatile("fnclex ; fwait");

}

static inline void restore_fpu(struct task_struct *tsk)

{

	/*

	 * The "nop" is needed to make the instructions the same

	 * length.

	 */

	alternative_input(

		"nop ; frstor %1",

		"fxrstor %1",

		X86_FEATURE_FXSR,

		"m" ((tsk)->thread.i387.fxsave));

}

/* We need a safe address that is cheap to find and that is already

   in L1 during context switch. The best choices are unfortunately

   different for UP and SMP */

#ifdef CONFIG_SMP

#define safe_address (__per_cpu_offset[0])

#else

#define safe_address (kstat_cpu(0).cpustat.user)

#endif

/*

 * These must be called with preempt disabled

 */

static inline void __save_init_fpu(struct task_struct *tsk)

{

	/* Use more nops than strictly needed in case the compiler

	   varies code */

	alternative_input(

		"fnsave %[fx] ;fwait;" GENERIC_NOP8 GENERIC_NOP4,

		"fxsave %[fx]\n"

		"bt $7,%[fsw] ; jnc 1f ; fnclex\n1:",

		X86_FEATURE_FXSR,

		[fx] "m" (tsk->thread.i387.fxsave),

		[fsw] "m" (tsk->thread.i387.fxsave.swd) : "memory");

	/* AMD K7/K8 CPUs don't save/restore FDP/FIP/FOP unless an exception

	   is pending.  Clear the x87 state here by setting it to fixed

	   values. safe_address is a random variable that should be in L1 */

	alternative_input(

		GENERIC_NOP8 GENERIC_NOP2,

		"emms\n\t"	  	/* clear stack tags */

		"fildl %[addr]", 	/* set F?P to defined value */

		X86_FEATURE_FXSAVE_LEAK,

		[addr] "m" (safe_address));

	task_thread_info(tsk)->status &= ~TS_USEDFPU;

}

/*

 * Signal frame handlers...

 */

extern int save_i387(struct _fpstate __user *buf);

extern int restore_i387(struct _fpstate __user *buf);

#endif	/* CONFIG_X86_64 */

static inline void __unlazy_fpu(struct task_struct *tsk)

{

	if (task_thread_info(tsk)->status & TS_USEDFPU) {

		__save_init_fpu(tsk);

		stts();

	} else

		tsk->fpu_counter = 0;

}

static inline void __clear_fpu(struct task_struct *tsk)

{

	if (task_thread_info(tsk)->status & TS_USEDFPU) {

		tolerant_fwait();

		task_thread_info(tsk)->status &= ~TS_USEDFPU;

		stts();

	}

}

static inline void kernel_fpu_begin(void)

{

	struct thread_info *me = current_thread_info();

	preempt_disable();

	if (me->status & TS_USEDFPU)

		__save_init_fpu(me->task);

	else

		clts();

}

static inline void kernel_fpu_end(void)

{

	stts();

	preempt_enable();

}

#ifdef CONFIG_X86_64

static inline void save_init_fpu(struct task_struct *tsk)

{

	__save_init_fpu(tsk);

	stts();

}

#define unlazy_fpu	__unlazy_fpu

#define clear_fpu	__clear_fpu

#else  /* CONFIG_X86_32 */

/*

 * These disable preemption on their own and are safe

 */

static inline void save_init_fpu(struct task_struct *tsk)

{

	preempt_disable();

	__save_init_fpu(tsk);

	stts();

	preempt_enable();

}

static inline void unlazy_fpu(struct task_struct *tsk)

{

	preempt_disable();

	__unlazy_fpu(tsk);

	preempt_enable();

}

static inline void clear_fpu(struct task_struct *tsk)

{

	preempt_disable();

	__clear_fpu(tsk);

	preempt_enable();

}

#endif	/* CONFIG_X86_64 */

/*

 * ptrace request handlers...

 */

extern int get_fpregs(struct user_i387_struct __user *buf,

		      struct task_struct *tsk);

extern int set_fpregs(struct task_struct *tsk,

		      struct user_i387_struct __user *buf);

struct user_fxsr_struct;

extern int get_fpxregs(struct user_fxsr_struct __user *buf,

		       struct task_struct *tsk);

extern int set_fpxregs(struct task_struct *tsk,

		       struct user_fxsr_struct __user *buf);

/*

 * i387 state interaction

 */

static inline unsigned short get_fpu_cwd(struct task_struct *tsk)

{

	if (cpu_has_fxsr) {

		return tsk->thread.i387.fxsave.cwd;

	} else {

		return (unsigned short)tsk->thread.i387.fsave.cwd;

	}

}

static inline unsigned short get_fpu_swd(struct task_struct *tsk)

{

	if (cpu_has_fxsr) {

		return tsk->thread.i387.fxsave.swd;

	} else {

		return (unsigned short)tsk->thread.i387.fsave.swd;

	}

}

static inline unsigned short get_fpu_mxcsr(struct task_struct *tsk)

{

	if (cpu_has_xmm) {

		return tsk->thread.i387.fxsave.mxcsr;

	} else {

		return MXCSR_DEFAULT;

	}

}

#endif	/* _ASM_X86_I387_H */

/*

 * Copyright (C) 1994 Linus Torvalds

 *

 * Pentium III FXSR, SSE support

 * General FPU state handling cleanups

 *	Gareth Hughes <gareth@valinux.com>, May 2000

 */

#ifndef __ASM_I386_I387_H

#define __ASM_I386_I387_H

#include <linux/sched.h>

#include <linux/init.h>

#include <linux/kernel_stat.h>

#include <asm/processor.h>

#include <asm/sigcontext.h>

#include <asm/user.h>

extern void mxcsr_feature_mask_init(void);

extern void init_fpu(struct task_struct *);

/*

 * FPU lazy state save handling...

 */

/*

 * The "nop" is needed to make the instructions the same

 * length.

 */

#define restore_fpu(tsk)			\

	alternative_input(			\

		"nop ; frstor %1",		\

		"fxrstor %1",			\

		X86_FEATURE_FXSR,		\

		"m" ((tsk)->thread.i387.fxsave))