* General FPU state handling cleanups
* Gareth Hughes <gareth@valinux.com>, May 2000
*/
-
-#include <linux/sched.h>
#include <linux/module.h>
#include <linux/regset.h>
+#include <linux/sched.h>
+
+#include <asm/sigcontext.h>
#include <asm/processor.h>
-#include <asm/i387.h>
#include <asm/math_emu.h>
-#include <asm/sigcontext.h>
-#include <asm/user.h>
-#include <asm/ptrace.h>
#include <asm/uaccess.h>
+#include <asm/ptrace.h>
+#include <asm/i387.h>
+#include <asm/user.h>
#ifdef CONFIG_X86_64
-
-#include <asm/sigcontext32.h>
-#include <asm/user32.h>
-
+# include <asm/sigcontext32.h>
+# include <asm/user32.h>
#else
-
-#define save_i387_ia32 save_i387
-#define restore_i387_ia32 restore_i387
-
-#define _fpstate_ia32 _fpstate
-#define user_i387_ia32_struct user_i387_struct
-#define user32_fxsr_struct user_fxsr_struct
-
+# define save_i387_ia32 save_i387
+# define restore_i387_ia32 restore_i387
+# define _fpstate_ia32 _fpstate
+# define user_i387_ia32_struct user_i387_struct
+# define user32_fxsr_struct user_fxsr_struct
#endif
#ifdef CONFIG_MATH_EMULATION
-#define HAVE_HWFP (boot_cpu_data.hard_math)
+# define HAVE_HWFP (boot_cpu_data.hard_math)
#else
-#define HAVE_HWFP 1
+# define HAVE_HWFP 1
#endif
-unsigned int mxcsr_feature_mask __read_mostly = 0xffffffffu;
+static unsigned int mxcsr_feature_mask __read_mostly = 0xffffffffu;
void mxcsr_feature_mask_init(void)
{
unsigned long mask = 0;
+
clts();
if (cpu_has_fxsr) {
memset(¤t->thread.i387.fxsave, 0,
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 */
+ write_cr0(oldcr0 & ~(X86_CR0_TS|X86_CR0_EM)); /* clear TS and EM */
mxcsr_feature_mask_init();
/* clean state in init */
if (!cpu_has_fxsr)
return -ENODEV;
- unlazy_fpu(target);
+ init_fpu(target);
return user_regset_copyout(&pos, &count, &kbuf, &ubuf,
&target->thread.i387.fxsave, 0, -1);
if (!cpu_has_fxsr)
return -ENODEV;
- unlazy_fpu(target);
+ init_fpu(target);
set_stopped_child_used_math(target);
ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
tmp = (tmp | (tmp >> 1)) & 0x3333; /* 00VV00VV00VV00VV */
tmp = (tmp | (tmp >> 2)) & 0x0f0f; /* 0000VVVV0000VVVV */
tmp = (tmp | (tmp >> 4)) & 0x00ff; /* 00000000VVVVVVVV */
+
return tmp;
}
* FXSR floating point environment conversions.
*/
-static void convert_from_fxsr(struct user_i387_ia32_struct *env,
- struct task_struct *tsk)
+static void
+convert_from_fxsr(struct user_i387_ia32_struct *env, struct task_struct *tsk)
{
struct i387_fxsave_struct *fxsave = &tsk->thread.i387.fxsave;
struct _fpreg *to = (struct _fpreg *) &env->st_space[0];
* should be actually ds/cs at fpu exception time, but
* that information is not available in 64bit mode.
*/
- asm("mov %%ds,%0" : "=r" (env->fos));
- asm("mov %%cs,%0" : "=r" (env->fcs));
+ asm("mov %%ds, %[fos]" : [fos] "=r" (env->fos));
+ asm("mov %%cs, %[fcs]" : [fcs] "=r" (env->fcs));
} else {
struct pt_regs *regs = task_pt_regs(tsk);
+
env->fos = 0xffff0000 | tsk->thread.ds;
env->fcs = regs->cs;
}
#else
env->fip = fxsave->fip;
- env->fcs = fxsave->fcs;
+ env->fcs = (u16) fxsave->fcs | ((u32) fxsave->fop << 16);
env->foo = fxsave->foo;
env->fos = fxsave->fos;
#endif
if (!HAVE_HWFP)
return fpregs_soft_get(target, regset, pos, count, kbuf, ubuf);
- unlazy_fpu(target);
+ init_fpu(target);
- if (!cpu_has_fxsr)
+ if (!cpu_has_fxsr) {
return user_regset_copyout(&pos, &count, &kbuf, &ubuf,
&target->thread.i387.fsave, 0, -1);
+ }
if (kbuf && pos == 0 && count == sizeof(env)) {
convert_from_fxsr(kbuf, target);
}
convert_from_fxsr(&env, target);
+
return user_regset_copyout(&pos, &count, &kbuf, &ubuf, &env, 0, -1);
}
if (!HAVE_HWFP)
return fpregs_soft_set(target, regset, pos, count, kbuf, ubuf);
- unlazy_fpu(target);
+ init_fpu(target);
set_stopped_child_used_math(target);
- if (!cpu_has_fxsr)
+ if (!cpu_has_fxsr) {
return user_regset_copyin(&pos, &count, &kbuf, &ubuf,
&target->thread.i387.fsave, 0, -1);
+ }
if (pos > 0 || count < sizeof(env))
convert_from_fxsr(&env, target);
{
if (!used_math())
return 0;
-
- /* This will cause a "finit" to be triggered by the next
+ /*
+ * This will cause a "finit" to be triggered by the next
* attempted FPU operation by the 'current' process.
*/
clear_used_math();
- if (HAVE_HWFP) {
- if (cpu_has_fxsr) {
- return save_i387_fxsave(buf);
- } else {
- return save_i387_fsave(buf);
- }
- } else {
+ if (!HAVE_HWFP) {
return fpregs_soft_get(current, NULL,
0, sizeof(struct user_i387_ia32_struct),
NULL, buf) ? -1 : 1;
}
+
+ if (cpu_has_fxsr)
+ return save_i387_fxsave(buf);
+ else
+ return save_i387_fsave(buf);
}
static inline int restore_i387_fsave(struct _fpstate_ia32 __user *buf)
{
struct task_struct *tsk = current;
+
clear_fpu(tsk);
return __copy_from_user(&tsk->thread.i387.fsave, buf,
sizeof(struct i387_fsave_struct));
static int restore_i387_fxsave(struct _fpstate_ia32 __user *buf)
{
- int err;
struct task_struct *tsk = current;
struct user_i387_ia32_struct env;
+ int err;
+
clear_fpu(tsk);
err = __copy_from_user(&tsk->thread.i387.fxsave, &buf->_fxsr_env[0],
sizeof(struct i387_fxsave_struct));
if (err || __copy_from_user(&env, buf, sizeof(env)))
return 1;
convert_to_fxsr(tsk, &env);
+
return 0;
}
int err;
if (HAVE_HWFP) {
- if (cpu_has_fxsr) {
+ if (cpu_has_fxsr)
err = restore_i387_fxsave(buf);
- } else {
+ else
err = restore_i387_fsave(buf);
- }
} else {
err = fpregs_soft_set(current, NULL,
0, sizeof(struct user_i387_ia32_struct),
NULL, buf) != 0;
}
set_used_math();
- return err;
-}
-
-#endif /* CONFIG_X86_32 || CONFIG_IA32_EMULATION */
-
-#ifdef CONFIG_X86_64
-
-int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *tsk)
-{
- return xfpregs_get(tsk, NULL, 0, sizeof(*buf), NULL, buf);
-}
-int set_fpregs(struct task_struct *tsk, struct user_i387_struct __user *buf)
-{
- return xfpregs_set(tsk, NULL, 0, sizeof(*buf), NULL, buf);
-}
-
-#else
-
-int get_fpregs(struct user_i387_struct __user *buf, struct task_struct *tsk)
-{
- return fpregs_get(tsk, NULL, 0, sizeof(*buf), NULL, buf);
-}
-
-int set_fpregs(struct task_struct *tsk, struct user_i387_struct __user *buf)
-{
- return fpregs_set(tsk, NULL, 0, sizeof(*buf), NULL, buf);
-}
-
-int get_fpxregs(struct user_fxsr_struct __user *buf, struct task_struct *tsk)
-{
- return xfpregs_get(tsk, NULL, 0, sizeof(*buf), NULL, buf);
-}
-
-int set_fpxregs(struct task_struct *tsk, struct user_fxsr_struct __user *buf)
-{
- return xfpregs_get(tsk, NULL, 0, sizeof(*buf), NULL, buf);
+ return err;
}
-#endif
-
/*
* FPU state for core dumps.
+ * This is only used for a.out dumps now.
+ * It is declared generically using elf_fpregset_t (which is
+ * struct user_i387_struct) but is in fact only used for 32-bit
+ * dumps, so on 64-bit it is really struct user_i387_ia32_struct.
*/
-
-static inline void copy_fpu_fsave(struct task_struct *tsk,
- struct user_i387_struct *fpu)
-{
- memcpy(fpu, &tsk->thread.i387.fsave,
- sizeof(struct user_i387_struct));
-}
-
-static inline void copy_fpu_fxsave(struct task_struct *tsk,
- struct user_i387_struct *fpu)
-{
- unsigned short *to;
- unsigned short *from;
- int i;
-
- memcpy(fpu, &tsk->thread.i387.fxsave, 7 * sizeof(long));
-
- to = (unsigned short *)&fpu->st_space[0];
- from = (unsigned short *)&tsk->thread.i387.fxsave.st_space[0];
- for (i = 0; i < 8; i++, to += 5, from += 8)
- memcpy(to, from, 5 * sizeof(unsigned short));
-}
-
int dump_fpu(struct pt_regs *regs, struct user_i387_struct *fpu)
{
- int fpvalid;
struct task_struct *tsk = current;
+ int fpvalid;
fpvalid = !!used_math();
- if (fpvalid) {
- unlazy_fpu(tsk);
- if (cpu_has_fxsr) {
- copy_fpu_fxsave(tsk, fpu);
- } else {
- copy_fpu_fsave(tsk, fpu);
- }
- }
+ if (fpvalid)
+ fpvalid = !fpregs_get(tsk, NULL,
+ 0, sizeof(struct user_i387_ia32_struct),
+ fpu, NULL);
return fpvalid;
}
EXPORT_SYMBOL(dump_fpu);
-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);
- if (cpu_has_fxsr)
- copy_fpu_fxsave(tsk, fpu);
- else
- copy_fpu_fsave(tsk, fpu);
- }
- return fpvalid;
-}
-
-int dump_task_extended_fpu(struct task_struct *tsk,
- struct user32_fxsr_struct *fpu)
-{
- int fpvalid = tsk_used_math(tsk) && cpu_has_fxsr;
-
- if (fpvalid) {
- if (tsk == current)
- unlazy_fpu(tsk);
- memcpy(fpu, &tsk->thread.i387.fxsave, sizeof(*fpu));
- }
- return fpvalid;
-}
+#endif /* CONFIG_X86_32 || CONFIG_IA32_EMULATION */