Merge with Linux 2.3.99-pre4.

16 files changed, 706 insertions, 744 deletions

diff --git a/arch/sh/Makefile b/arch/sh/Makefile
index 83591c86c..1857ef33b 100644
--- a/arch/sh/Makefile
+++ b/arch/sh/Makefile
@@ -1,4 +1,4 @@
-# $Id: Makefile,v 1.2 1999/12/23 12:13:53 gniibe Exp gniibe $
+# $Id: Makefile,v 1.4 2000/03/08 15:14:14 gniibe Exp $
 #
 # This file is subject to the terms and conditions of the GNU General Public
 # License.  See the file "COPYING" in the main directory of this archive
diff --git a/arch/sh/config.in b/arch/sh/config.in
index e3bfd113c..2ee913c3c 100644
--- a/arch/sh/config.in
+++ b/arch/sh/config.in
@@ -48,6 +48,7 @@ mainmenu_option next_comment
 comment 'General setup'
 
 define_bool CONFIG_ISA n
+define_bool CONFIG_SBUS n
 
 bool 'Networking support' CONFIG_NET
 
@@ -73,6 +74,8 @@ bool 'Support for hot-pluggable devices' CONFIG_HOTPLUG
 
 if [ "$CONFIG_HOTPLUG" = "y" ] ; then
    source drivers/pcmcia/Config.in
+else
+   define_bool CONFIG_PCMCIA n
 fi
 
 bool 'System V IPC' CONFIG_SYSVIPC
@@ -170,8 +173,6 @@ if [ "$CONFIG_HOTPLUG" = "y" -a "$CONFIG_PCMCIA" != "n" ]; then
    source drivers/char/pcmcia/Config.in
 fi
 
-#source drivers/misc/Config.in
-
 source fs/Config.in
 
 if [ "$CONFIG_VT" = "y" ]; then
diff --git a/arch/sh/defconfig b/arch/sh/defconfig
index fad2ab8b1..337e46956 100644
--- a/arch/sh/defconfig
+++ b/arch/sh/defconfig
@@ -29,10 +29,12 @@ CONFIG_MEMORY_START=0c000000
 # General setup
 #
 # CONFIG_ISA is not set
+# CONFIG_SBUS is not set
 # CONFIG_NET is not set
 CONFIG_CF_ENABLER=y
 # CONFIG_PCI is not set
 # CONFIG_HOTPLUG is not set
+# CONFIG_PCMCIA is not set
 # CONFIG_SYSVIPC is not set
 # CONFIG_BSD_PROCESS_ACCT is not set
 # CONFIG_SYSCTL is not set
diff --git a/arch/sh/kernel/Makefile b/arch/sh/kernel/Makefile
index efa2fb109..8996a13dc 100644
--- a/arch/sh/kernel/Makefile
+++ b/arch/sh/kernel/Makefile
@@ -7,7 +7,7 @@
 #
 
 .S.o:
-	$(CC) -D__ASSEMBLY__ $(AFLAGS) -traditional -c $< -o $*.o
+	$(CC) $(AFLAGS) -traditional -c $< -o $*.o
 
 O_TARGET := kernel.o
 O_OBJS   := process.o signal.o entry.o traps.o irq.o irq_onchip.o \
@@ -29,7 +29,7 @@ all: kernel.o head.o init_task.o
 entry.o: entry.S
 
 head.o: head.S
-	$(CC) -D__ASSEMBLY__ $(AFLAGS) -traditional -c $*.S -o $*.o
+	$(CC) $(AFLAGS) -traditional -c $*.S -o $*.o
 
 clean:
 
diff --git a/arch/sh/kernel/entry.S b/arch/sh/kernel/entry.S
index 77bb7938b..46ad20f53 100644
--- a/arch/sh/kernel/entry.S
+++ b/arch/sh/kernel/entry.S
@@ -1,4 +1,4 @@
-/* $Id: entry.S,v 1.55 2000/03/05 01:48:58 gniibe Exp $
+/* $Id: entry.S,v 1.71 2000/03/22 13:29:33 gniibe Exp $
  *
  *  linux/arch/sh/entry.S
  *
@@ -33,10 +33,9 @@
  *
  *	syscall #
  *	ssr
- *	r15 = stack pointer
  *	r0
  *      ...
- *	r14
+ *	r15 = stack pointer
  *	gbr
  *	mach
  *	macl
@@ -46,7 +45,7 @@
  */
 
 /*
- * these are offsets into the task-struct.
+ * These are offsets into the task-struct.
  */
 state		=  0
 flags		=  4
@@ -78,8 +77,8 @@ MMU_TEA = 0xff00000c		! TLB Exception Address Register
 /* Offsets to the stack */
 SYSCALL_NR =  0
 SR	   =  4
-SP	   =  8
-R0         =  12
+R0         =  8
+SP	   =  (8+15*4)
 
 #define k0	r0
 #define k1	r1
@@ -96,7 +95,7 @@ R0         =  12
 	k2	scratch (Exception code)
 	k3	scratch (Return address)
 	k4	Stack base = current+8192
-	k5	reserved
+	k5	Global Interrupt Mask (0--15)
 	k6	reserved
 	k7	reserved
 */
@@ -115,109 +114,113 @@ R0         =  12
 ! this first version depends *much* on C implementation.
 !
 
-#define DO_FAULT(write)			\
-	mov.l	4f,r0;			\
-	mov.l	@r0,r6; 		\
-	/* STI */ 			\
-	mov.l	3f,r1; 			\
-	stc	sr,r0; 			\
-	and	r1,r0; 			\
-	ldc	r0,sr; 			\
-	/*    */ 			\
-	mov	r15,r4; 		\
-	mov.l	2f,r0; 			\
-	jmp	@r0; 			\
-	 mov	#write,r5;
+#define RESTORE_FLAGS()			\
+	mov.l	@(SR,$r15), $r0;	\
+	and	#0xf0, $r0;		\
+	shlr8	$r0;			\
+	cmp/eq	#0x0f, $r0;		\
+	bt	9f;			\
+	mov.l	__INV_IMASK, $r1;	\
+	stc	$sr, $r0;		\
+	and	$r1, $r0;		\
+	stc	$r5_bank, $r1;		\
+	or	$r1, $r0;		\
+	ldc	$r0, $sr
 
 	.balign	4
 tlb_protection_violation_load:
 tlb_miss_load:
-	mov	#-1,r0
-	mov.l	r0,@r15		! syscall nr = -1
-	DO_FAULT(0)
+	mov	#-1, $r0
+	mov.l	$r0, @$r15	! syscall nr = -1
+	mov.l	2f, $r0
+	mov.l	@$r0, $r6
+	RESTORE_FLAGS()
+9:	mov	$r15, $r4
+	mov.l	1f, $r0
+	jmp	@$r0
+	 mov	#0, $r5
 
 	.balign	4
 tlb_protection_violation_store:
 tlb_miss_store:
 initial_page_write:
-	mov	#-1,r0
-	mov.l	r0,@r15		! syscall nr = -1
-	DO_FAULT(1)
+	mov	#-1, $r0
+	mov.l	$r0, @$r15	! syscall nr = -1
+	mov.l	2f, $r0
+	mov.l	@$r0, $r6
+	RESTORE_FLAGS()
+9:	mov	$r15, $r4
+	mov.l	1f, $r0
+	jmp	@$r0
+	 mov	#1, $r5
 
 	.balign 4
-2:	.long	SYMBOL_NAME(do_page_fault)
-3:	.long	0xefffffff	! BL=0
-4:	.long	MMU_TEA
+1:	.long	SYMBOL_NAME(do_page_fault)
+2:	.long	MMU_TEA
 
 #ifdef CONFIG_DEBUG_KERNEL_WITH_GDB_STUB
 	.balign	4
 	/* Unwind the stack and jmp to the debug entry */
 debug:
-	add	#4,r15		! skip syscall number
-	mov.l	@r15+,r11	! SSR
-	mov.l	@r15+,r10	! original stack
-	mov.l	@r15+,r0
-	mov.l	@r15+,r1
-	mov.l	@r15+,r2
-	mov.l	@r15+,r3
-	mov.l	@r15+,r4
-	mov.l	@r15+,r5
-	mov.l	@r15+,r6
-	mov.l	@r15+,r7
-	stc	sr,r14
-	mov.l	8f,r9			! BL =1, RB=1
-	or	r9,r14
-	ldc	r14,sr			! here, change the register bank
-	mov	r10,k0
-	mov	r11,k1
-	mov.l	@r15+,r8
-	mov.l	@r15+,r9
-	mov.l	@r15+,r10
-	mov.l	@r15+,r11
-	mov.l	@r15+,r12
-	mov.l	@r15+,r13
-	mov.l	@r15+,r14
-	ldc.l	@r15+,gbr
-	lds.l	@r15+,mach
-	lds.l	@r15+,macl
-	lds.l	@r15+,pr
-	ldc.l	@r15+,spc
-	mov	k0,r15
+	add	#4, $r15	! skip syscall number
+	mov.l	@$r15+, $r11	! SSR
+	mov.l	@$r15+, $r0
+	mov.l	@$r15+, $r1
+	mov.l	@$r15+, $r2
+	mov.l	@$r15+, $r3
+	mov.l	@$r15+, $r4
+	mov.l	@$r15+, $r5
+	mov.l	@$r15+, $r6
+	mov.l	@$r15+, $r7
+	stc	$sr, $r14
+	mov.l	1f, $r9			! BL =1, RB=1
+	or	$r9, $r14
+	ldc	$r14, $sr		! here, change the register bank
+	mov	$r11, $k1
+	mov.l	@$r15+, $r8
+	mov.l	@$r15+, $r9
+	mov.l	@$r15+, $r10
+	mov.l	@$r15+, $r11
+	mov.l	@$r15+, $r12
+	mov.l	@$r15+, $r13
+	mov.l	@$r15+, $r14
+	mov.l	@$r15+, $k0
+	ldc.l	@$r15+, $gbr
+	lds.l	@$r15+, $mach
+	lds.l	@$r15+, $macl
+	lds.l	@$r15+, $pr
+	ldc.l	@$r15+, $spc
+	mov	$k0, $r15
 	!
-	mov.l	9f,k0
-	jmp	@k0
-	 ldc	k1,ssr
+	mov.l	2f, $k0
+	jmp	@$k0
+	 ldc	$k1, $ssr
 	.balign	4
-8:	.long	0x300000f0
-9:	.long	0xa0000100
+1:	.long	0x300000f0
+2:	.long	0xa0000100
 #endif
 
 	.balign	4
 error:	
-	! STI
-	mov.l	2f,r1
-	stc	sr,r0
-	and	r1,r0
-	ldc	r0,sr
 	!
-	mov.l	1f,r1
-	mov	#-1,r0
-	jmp	@r1
-	 mov.l	r0,@r15		! syscall nr = -1
+	RESTORE_FLAGS()
+9:	mov.l	1f, $r1
+	mov	#-1, $r0
+	jmp	@$r1
+	 mov.l	$r0, @$r15		! syscall nr = -1
 	.balign	4
 1:	.long	SYMBOL_NAME(do_exception_error)
-2:	.long	0xefffffff	! BL=0
 
-badsys:	mov	#-ENOSYS,r0
+badsys:	mov	#-ENOSYS, $r0
 	rts			! go to ret_from_syscall..
-	 mov.l	r0,@(R0,r15)
+	 mov.l	$r0, @(R0,$r15)
 
 !
 !
 !
 ENTRY(ret_from_fork)
 	bra	SYMBOL_NAME(ret_from_syscall)
-	 add	#4,r15		! pop down bogus r0 (see switch_to MACRO)
+	 add	#4, $r15	! pop down bogus r0 (see switch_to MACRO)
 
 !
 ! The immediate value of "trapa" indicates the number of arguments
@@ -226,83 +229,77 @@ ENTRY(ret_from_fork)
 ! Note that TRA register contains the value = Imm x 4.
 !
 system_call:
-	mov.l	1f,r2
-	mov.l	@r2,r8
+	mov.l	1f, $r2
+	mov.l	@$r2, $r8
 	!
 	! DEBUG DEBUG
-	! mov.l	led,r1
-	! mov	r0,r2
-	! mov.b	r2,@r1
+	! mov.l	led, $r1
+	! mov	$r0, $r2
+	! mov.b	$r2, @$r1
 	!
 #ifdef CONFIG_DEBUG_KERNEL_WITH_GDB_STUB
-	mov	#0x20,r1
-	extu.b	r1,r1
-	shll2	r1
-	cmp/hs	r1,r8
+	mov	#0x20, $r1
+	extu.b	$r1, $r1
+	shll2	$r1
+	cmp/hs	$r1, $r8
 	bt	debug
 #endif
-	! STI
-	mov.l	2f,r1
-	stc	sr,r2
-	and	r1,r2
-	ldc	r2,sr
 	!
-	mov.l	__n_sys,r1
-	cmp/hs	r1,r0
-	bt/s	badsys
-	 mov	r0,r2
+	mov	$r0, $r2
+	RESTORE_FLAGS()
+9:	mov.l	__n_sys, $r1
+	cmp/hs	$r1, $r2
+	bt	badsys
 	!
-	stc	ksp,r1		!
-	mov.l	__tsk_flags,r0	!
-	add	r0,r1		!
-	mov.l	@r1,r0		! Is it trace?
-	tst	#PF_TRACESYS,r0
+	stc	$ksp, $r1
+	mov.l	__tsk_flags, $r0
+	add	$r0, $r1	!
+	mov.l	@$r1, $r0	! Is it trace?
+	tst	#PF_TRACESYS, $r0
 	bt	5f
 	!                     Trace system call
-	mov	#-ENOSYS,r1
-	mov.l	r1,@(R0,r15)
-	mov.l	3f,r1
-	jsr	@r1
+	mov	#-ENOSYS, $r1
+	mov.l	$r1, @(R0,$r15)
+	mov.l	3f, $r1
+	jsr	@$r1
 	 nop
-	mova	4f,r0
+	mova	3f, $r0
 	bra	6f
-	 lds	r0,pr
-	!
-5:	mova	ret,r0		! normal case
-	lds	r0,pr
-	!			Build the stack frame if TRA > 0
+	 lds	$r0, $pr
 	!
-6:	mov	r2,r3
-	mov	r8,r2
-	cmp/pl	r8
-	bf	9f
-	mov.l	@(SP,r15),r0	! get original stack
-7:	add	#-4,r8
-8:	mov.l	@(r0,r8),r1	! May cause address error exception..
-	mov.l	r1,@-r15
-	cmp/pl	r8
+5:	mova	syscall_ret, $r0
+	lds	$r0, $pr
+	!				  Build the stack frame if TRA > 0
+6:	mov	$r2, $r3
+	mov	$r8, $r2
+	cmp/pl	$r8
+	bf	0f
+	mov	#SP, $r0
+	mov.l	@($r0,$r15), $r0	! get original stack
+7:	add	#-4, $r8
+4:	mov.l	@($r0,$r8), $r1		! May cause address error exception..
+	mov.l	$r1, @-$r15
+	cmp/pl	$r8
 	bt	7b
 	!
-9:	mov	r3,r0
-	shll2	r0		! x4
-	mov.l	__sct,r1
-	add	r1,r0
-	mov.l	@r0,r1
-	jmp	@r1
-	 mov	r2,r8
-
+0:	mov	$r3, $r0
+	shll2	$r0		! x4
+	mov.l	__sct, $r1
+	add	$r1, $r0
+	mov.l	@$r0, $r1
+	jmp	@$r1
+	 mov	$r2, $r8
 	! In case of trace
 	.balign	4
-4:	add	r8,r15		! pop off the arguments
-	mov.l	r0,@(R0,r15)	! save the return value
-	mov.l	3f,r1
-	mova	SYMBOL_NAME(ret_from_syscall),r0
-	jmp	@r1
-	 lds	r0,pr
+3:	add	$r8, $r15		! pop off the arguments
+	mov.l	$r0, @(R0,$r15)		! save the return value
+	mov.l	2f, $r1
+	mova	SYMBOL_NAME(ret_from_syscall), $r0
+	jmp	@$r1
+	 lds	$r0, $pr
 	.balign	4
-3:	.long	SYMBOL_NAME(syscall_trace)
-2:	.long	0xefffffff	! BL=0
 1:	.long	TRA
+2:	.long	SYMBOL_NAME(syscall_trace)
 __n_sys:	.long	NR_syscalls
 __sct:		.long	SYMBOL_NAME(sys_call_table)
 __tsk_flags:	.long	flags-8192	! offset from stackbase to tsk->flags
@@ -311,85 +308,80 @@ led:	.long	0xa8000000	! For my board -- gN
 	.section	.fixup,"ax"
 fixup_syscall_argerr:
 	rts
-	 mov.l	1f,r0
+	 mov.l	1f, $r0
 1:	.long	-22	! -EINVAL
 .previous
 
 	.section	__ex_table, "a"
 	.balign	4
-	.long	8b,fixup_syscall_argerr
+	.long	4b,fixup_syscall_argerr
 .previous
 
+	.balign	4
 reschedule:
-	mova	SYMBOL_NAME(ret_from_syscall),r0
-	mov.l	1f,r1
-	jmp	@r1
-	 lds	r0,pr
+	mova	SYMBOL_NAME(ret_from_syscall), $r0
+	mov.l	1f, $r1
+	jmp	@$r1
+	 lds	$r0, $pr
 	.balign	4
 1:	.long	SYMBOL_NAME(schedule)
 
 ENTRY(ret_from_irq)
-	mov.l	@(SR,r15),r0	! get status register
-	shll	r0
-	shll	r0		! kernel space?
+	mov.l	@(SR,$r15), $r0	! get status register
+	shll	$r0
+	shll	$r0		! kernel space?
 	bt	restore_all	! Yes, it's from kernel, go back soon
-	! STI
-	mov.l	1f, $r1
-	stc	$sr, $r2
-	and	$r1, $r2
-	ldc	$r2, $sr
 	!
-	bra	ret_with_reschedule
+	RESTORE_FLAGS()
+9:	bra	ret_with_reschedule
 	 nop
 
 ENTRY(ret_from_exception)
-	mov.l	@(SR,r15),r0	! get status register
-	shll	r0
-	shll	r0		! kernel space?
+	mov.l	@(SR,$r15), $r0	! get status register
+	shll	$r0
+	shll	$r0		! kernel space?
 	bt	restore_all	! Yes, it's from kernel, go back soon
-	! STI
-	mov.l	1f, $r1
-	stc	$sr, $r2
-	and	$r1, $r2
-	ldc	$r2, $sr
 	!
-	bra	ret_from_syscall
+	RESTORE_FLAGS()
+9:	bra	ret_from_syscall
 	 nop
 	.balign	4
-1:	.long	0xefffffff	! BL=0
+__INV_IMASK:
+	.long	0xffffff0f	! ~(IMASK)
 
 	.balign 4
-ret:	add	r8,r15		! pop off the arguments
-	mov.l	r0,@(R0,r15)	! save the return value
+syscall_ret:
+	add	$r8, $r15	! pop off the arguments
+	mov.l	$r0, @(R0,$r15)	! save the return value
 	/* fall through */
 
 ENTRY(ret_from_syscall)
-	mov.l	__softirq_state,r0
-	mov.l	@r0,r1
-	mov.l	@(4,r0),r2
-	tst	r2,r1
+	mov.l	__softirq_state, $r0
+	mov.l	@$r0, $r1
+	mov.l	@(4,$r0), $r2
+	tst	$r2, $r1
 	bt	ret_with_reschedule
 handle_softirq:
-	mov.l	__do_softirq,r0
-	jsr	@r0
+	mov.l	__do_softirq, $r0
+	jsr	@$r0
 	 nop
 ret_with_reschedule:
-	stc	ksp,r1
-	mov.l	__minus8192,r0
-	add	r0,r1
-	mov.l	@(need_resched,r1),r0
-	tst	#0xff,r0
+	stc	$ksp, $r1
+	mov.l	__minus8192, $r0
+	add	$r0, $r1
+	mov.l	@(need_resched,$r1), $r0
+	tst	#0xff, $r0
 	bf	reschedule
-	mov.l	@(sigpending,r1),r0
-	tst	#0xff,r0
+	mov.l	@(sigpending,$r1), $r0
+	tst	#0xff, $r0
 	bt	restore_all
 signal_return:
-	mov	r15,r4
-	mov	#0,r5
-	mov.l	__do_signal,r1
-	mova	restore_all,r0
-	jmp	@r1
-	 lds	r0,pr
+	mov	$r15, $r4
+	mov	#0, $r5
+	mov.l	__do_signal, $r1
+	mova	restore_all, $r0
+	jmp	@$r1
+	 lds	$r0, $pr
 	.balign	4
 __do_signal:
 	.long	SYMBOL_NAME(do_signal)
@@ -407,56 +399,57 @@ restore_all:
 	jsr	@$r1
 	 stc	$sr, $r4
 #endif
-	add	#4,r15		! Skip syscall number
-	mov.l	@r15+,r11	! Got SSR into R11
+	add	#4, $r15	! Skip syscall number
+	mov.l	@$r15+, $r11	! Got SSR into R11
 #if defined(__SH4__)
 	mov	$r11, $r12
 #endif
 	!
-	mov.l	1f,r1
-	stc	sr,r0
-	and	r1,r0		! Get IMASK+FD
-	mov.l	2f,r1
-	and	r1,r11
-	or	r0,r11		! Inherit the IMASK+FD value of SR
+	mov.l	1f, $r1
+	stc	$sr, $r0
+	and	$r1, $r0	! Get FD
+	mov.l	2f, $r1
+	and	$r1, $r11
+	or	$r0, $r11	! Inherit the FD value of SR
+	stc	$r5_bank, $r0
+	or	$r0, $r11	! Inherit the IMASK value
 	!
-	mov.l	@r15+,r10	! original stack
-	mov.l	@r15+,r0
-	mov.l	@r15+,r1
-	mov.l	@r15+,r2
-	mov.l	@r15+,r3
-	mov.l	@r15+,r4
-	mov.l	@r15+,r5
-	mov.l	@r15+,r6
-	mov.l	@r15+,r7
-	stc	sr,r14
-	mov.l	__blrb_flags,r9		! BL =1, RB=1
-	or	r9,r14
-	ldc	r14,sr			! here, change the register bank
-	mov	r10,k0
-	mov	r11,k1
+	mov.l	@$r15+, $r0
+	mov.l	@$r15+, $r1
+	mov.l	@$r15+, $r2
+	mov.l	@$r15+, $r3
+	mov.l	@$r15+, $r4
+	mov.l	@$r15+, $r5
+	mov.l	@$r15+, $r6
+	mov.l	@$r15+, $r7
+	stc	$sr, $r14
+	mov.l	__blrb_flags, $r9	! BL =1, RB=1
+	or	$r9, $r14
+	ldc	$r14, $sr		! here, change the register bank
+	mov	$r11, $k1
 #if defined(__SH4__)
 	mov	$r12, $k2
 #endif
-	mov.l	@r15+,r8
-	mov.l	@r15+,r9
-	mov.l	@r15+,r10
-	mov.l	@r15+,r11
-	mov.l	@r15+,r12
-	mov.l	@r15+,r13
-	mov.l	@r15+,r14
-	ldc.l	@r15+,gbr
-	lds.l	@r15+,mach
-	lds.l	@r15+,macl
-	lds.l	@r15+,pr
-	ldc.l	@r15+,spc
-	ldc	k1,ssr
+	mov.l	@$r15+, $r8
+	mov.l	@$r15+, $r9
+	mov.l	@$r15+, $r10
+	mov.l	@$r15+, $r11
+	mov.l	@$r15+, $r12
+	mov.l	@$r15+, $r13
+	mov.l	@$r15+, $r14
+	mov.l	@$r15+, $k0	! original stack
+	ldc.l	@$r15+, $gbr
+	lds.l	@$r15+, $mach
+	lds.l	@$r15+, $macl
+	lds.l	@$r15+, $pr
+	ldc.l	@$r15+, $spc
+	ldc	$k1, $ssr
 #if defined(__SH4__)
 	shll	$k1
 	shll	$k1
 	bf	9f		! user mode
 	/* Kernel to kernel transition */
-	mov.l	3f, $k1
+	mov.l	1f, $k1
 	tst	$k1, $k2
 	bf	9f		! it hadn't FPU
 	! Kernel to kernel and FPU was used
@@ -496,7 +489,7 @@ restore_all:
 	lds.l	@$r15+, $fpul
 9:
 #endif
-	mov	k0,r15
+	mov	$k0, $r15
 	rte
 	 nop
 
@@ -510,9 +503,8 @@ __init_task_flags:
 __PF_USEDFPU:
 	.long	PF_USEDFPU
 #endif
-1:	.long	0x000080f0	! IMASK+FD
+1:	.long	0x00008000	! FD
 2:	.long	0xffff7f0f	! ~(IMASK+FD)
-3:	.long	0x00008000	! FD=1
 
 ! Exception Vector Base
 !
@@ -524,10 +516,10 @@ ENTRY(vbr_base)
 !
 	.balign 	256,0,256
 general_exception:
-	mov.l	1f,k2
-	mov.l	2f,k3
+	mov.l	1f, $k2
+	mov.l	2f, $k3
 	bra	handle_exception
-	 mov.l	@k2,k2
+	 mov.l	@$k2, $k2
 	.balign	4
 2:	.long	SYMBOL_NAME(ret_from_exception)
 1:	.long	EXPEVT
@@ -535,17 +527,17 @@ general_exception:
 !
 	.balign 	1024,0,1024
 tlb_miss:
-	mov.l	1f,k2
-	mov.l	4f,k3
+	mov.l	1f, $k2
+	mov.l	4f, $k3
 	bra	handle_exception
-	 mov.l	@k2,k2
+	 mov.l	@$k2, $k2
 !
 	.balign 	512,0,512
 interrupt:
-	mov.l	2f,k2
-	mov.l	3f,k3
+	mov.l	2f, $k2
+	mov.l	3f, $k3
 	bra	handle_exception
-	 mov.l	@k2,k2
+	 mov.l	@$k2, $k2
 
 	.balign	4
 1:	.long	EXPEVT
@@ -559,9 +551,9 @@ handle_exception:
 	! Using k0, k1 for scratch registers (r0_bank1, r1_bank),
 	! save all registers onto stack.
 	!
-	stc	ssr,k0		! from kernel space?
-	shll	k0		! Check MD bit (bit30)
-	shll	k0
+	stc	$ssr, $k0	! from kernel space?
+	shll	$k0		! Check MD bit (bit30)
+	shll	$k0
 #if defined(__SH4__)
 	bf/s	8f		! it's from user to kernel transition
 	 mov	$r15, $k0	! save original stack to k0
@@ -569,6 +561,7 @@ handle_exception:
 	mov.l	2f, $k1
 	stc	$ssr, $k0
 	tst	$k1, $k0
+	mov.l	4f, $k1
 	bf/s	9f		! FPU is not used
 	 mov	$r15, $k0	! save original stack to k0
 	! FPU is used, save FPU
@@ -593,64 +586,63 @@ handle_exception:
 	fmov.s	$fr1, @-$r15
 	fmov.s	$fr0, @-$r15
 	bra 9f
-	 mov	#0, $k1
+	 mov.l	3f, $k1
 #else
+	mov.l	3f, $k1
 	bt/s	9f		! it's from kernel to kernel transition
-	 mov	r15,k0		! save original stack to k0 anyway
+	 mov	$r15, $k0	! save original stack to k0 anyway
 #endif
 8:	/* User space to kernel */
-	mov	kernel_sp, $r15	! change to kernel stack
-#if defined(__SH4__)
-	mov.l	2f, $k1		! let kernel release FPU
-#endif
-9:	stc.l	spc,@-r15
-	sts.l	pr,@-r15
+	mov	$kernel_sp, $r15	! change to kernel stack
+	mov.l	4f, $k1			! let kernel release FPU
+9:	stc.l	$spc, @-$r15
+	sts.l	$pr, @-$r15
 	!
-	lds	k3,pr		! Set the return address to pr
+	lds	$k3, $pr		! Set the return address to pr
 	!
-	sts.l	macl,@-r15
-	sts.l	mach,@-r15
-	stc.l	gbr,@-r15
-	mov.l	r14,@-r15
+	sts.l	$macl, @-$r15
+	sts.l	$mach, @-$r15
+	stc.l	$gbr, @-$r15
+	mov.l	$k0, @-$r15	! save orignal stack
+	mov.l	$r14, @-$r15
 	!
-	stc	sr,r14		! Back to normal register bank, and
-#if defined(__SH4__)
-	or	$k1, $r14	! may release FPU
-#endif
-	mov.l	3f,k1
-	and	k1,r14		! ...
-	ldc	r14,sr		! ...changed here.
+	stc	$sr, $r14	! Back to normal register bank, and
+	or	$k1, $r14	! Block all interrupts, may release FPU
+	mov.l	5f, $k1
+	and	$k1, $r14	! ...
+	ldc	$r14, $sr	! ...changed here.
 	!
-	mov.l	r13,@-r15
-	mov.l	r12,@-r15
-	mov.l	r11,@-r15
-	mov.l	r10,@-r15
-	mov.l	r9,@-r15
-	mov.l	r8,@-r15
-	mov.l	r7,@-r15
-	mov.l	r6,@-r15
-	mov.l	r5,@-r15
-	mov.l	r4,@-r15
-	mov.l	r3,@-r15
-	mov.l	r2,@-r15
-	mov.l	r1,@-r15
-	mov.l	r0,@-r15
-	stc.l	r0_bank,@-r15	! save orignal stack
-	stc.l	ssr,@-r15
-	mov.l	r0,@-r15	! push r0 again (for syscall number)
+	mov.l	$r13, @-$r15
+	mov.l	$r12, @-$r15
+	mov.l	$r11, @-$r15
+	mov.l	$r10, @-$r15
+	mov.l	$r9, @-$r15
+	mov.l	$r8, @-$r15
+	mov.l	$r7, @-$r15
+	mov.l	$r6, @-$r15
+	mov.l	$r5, @-$r15
+	mov.l	$r4, @-$r15
+	mov.l	$r3, @-$r15
+	mov.l	$r2, @-$r15
+	mov.l	$r1, @-$r15
+	mov.l	$r0, @-$r15
+	stc.l	$ssr, @-$r15
+	mov.l	$r0, @-$r15	! push $r0 again (for syscall number)
 	! Then, dispatch to the handler, according to the excepiton code.
-	stc	k_ex_code,r1
-	shlr2	r1
-	shlr	r1
-	mov.l	1f,r0
-	add	r1,r0
-	mov.l	@r0,r0
-	jmp	@r0
-	 mov.l	@r15,r0		! recovering r0..
+	stc	$k_ex_code, $r1
+	shlr2	$r1
+	shlr	$r1
+	mov.l	1f, $r0
+	add	$r1, $r0
+	mov.l	@$r0, $r0
+	jmp	@$r0
+	 mov.l	@$r15, $r0	! recovering $r0..
 	.balign	4
 1:	.long	SYMBOL_NAME(exception_handling_table)
 2:	.long	0x00008000	! FD=1
-3:	.long	0xdfffffff	! RB=0, leave BL=1
+3:	.long	0x000000f0	! FD=0, IMASK=15
+4:	.long	0x000080f0	! FD=1, IMASK=15
+5:	.long	0xcfffffff	! RB=0, BL=0
 
 none:
 	rts
@@ -679,7 +671,7 @@ ENTRY(exception_handling_table)
 ENTRY(nmi_slot)
 	.long	none				! Not implemented yet
 ENTRY(user_break_point_trap)
-	.long	error				! Not implemented yet
+	.long	break_point_trap
 ENTRY(interrupt_table)
 	! external hardware
 	.long	SYMBOL_NAME(do_IRQ)	! 0000
@@ -985,6 +977,8 @@ ENTRY(sys_call_table)
 	.long SYMBOL_NAME(sys_setfsuid)		/* 215 */
 	.long SYMBOL_NAME(sys_setfsgid)
 	.long SYMBOL_NAME(sys_pivot_root)
+	.long SYMBOL_NAME(sys_mincore)
+	.long SYMBOL_NAME(sys_madvise)
 
 	/*
 	 * NOTE!! This doesn't have to be exact - we just have
@@ -992,7 +986,7 @@ ENTRY(sys_call_table)
 	 * entries. Don't panic if you notice that this hasn't
 	 * been shrunk every time we add a new system call.
 	 */
-	.rept NR_syscalls-217
+	.rept NR_syscalls-219
 		.long SYMBOL_NAME(sys_ni_syscall)
 	.endr
 
diff --git a/arch/sh/kernel/fpu.c b/arch/sh/kernel/fpu.c
index 335902c1d..5301a1333 100644
--- a/arch/sh/kernel/fpu.c
+++ b/arch/sh/kernel/fpu.c
@@ -1,4 +1,4 @@
-/* $Id: fpu.c,v 1.27 2000/03/05 01:48:34 gniibe Exp $
+/* $Id: fpu.c,v 1.29 2000/03/22 13:42:10 gniibe Exp $
  *
  * linux/arch/sh/kernel/fpu.c
  *
@@ -114,8 +114,6 @@ restore_fpu(struct task_struct *tsk)
  * has the property that no matter wether considered as single or as
  * double precission represents signaling NANS.  
  */
-/* Double presision, NANS as NANS, rounding to nearest, no exceptions */
-#define FPU_DEFAULT  0x00080000
 
 void fpu_init(void)
 {
@@ -156,7 +154,7 @@ void fpu_init(void)
 		     "fsts	$fpul, $fr15\n\t"
 		     "frchg"
 		     : /* no output */
-		     : "r" (0), "r" (FPU_DEFAULT));
+		     : "r" (0), "r" (FPSCR_INIT));
 }
 
 asmlinkage void
diff --git a/arch/sh/kernel/head.S b/arch/sh/kernel/head.S
index 3f938557a..ae406d22d 100644
--- a/arch/sh/kernel/head.S
+++ b/arch/sh/kernel/head.S
@@ -1,4 +1,4 @@
-/* $Id: head.S,v 1.16 2000/03/02 00:01:15 gniibe Exp $
+/* $Id: head.S,v 1.17 2000/03/06 12:44:24 gniibe Exp $
  *
  *  arch/sh/kernel/head.S
  *
@@ -37,8 +37,11 @@ ENTRY(empty_zero_page)
  */
 ENTRY(_stext)
 	!			Initialize Status Register
-	mov.l	1f, $r0		! MD=1, RB=0, BL=1
+	mov.l	1f, $r0		! MD=1, RB=0, BL=0, IMASK=0xF
 	ldc	$r0, $sr
+	!			Initialize global interrupt mask
+	mov	#0, $r0
+	ldc	$r0, $r5_bank
 	!
 	mov.l	2f, $r0
 	mov	$r0, $r15	! Set initial r15 (stack pointer)
@@ -62,7 +65,7 @@ ENTRY(_stext)
 	 nop
 
 	.balign 4
-1:	.long	0x50000000		! MD=1, RB=0, BL=1, FD=0
+1:	.long	0x400000F0		! MD=1, RB=0, BL=0, FD=0, IMASK=0xF
 2:	.long	SYMBOL_NAME(stack)
 3:	.long	SYMBOL_NAME(__bss_start)
 4:	.long	SYMBOL_NAME(_end)
diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c
index a15352389..af03ef46d 100644
--- a/arch/sh/kernel/irq.c
+++ b/arch/sh/kernel/irq.c
@@ -1,4 +1,4 @@
-/* $Id: irq.c,v 1.11 2000/02/29 11:03:40 gniibe Exp $
+/* $Id: irq.c,v 1.12 2000/03/06 14:07:50 gniibe Exp $
  *
  * linux/arch/sh/kernel/irq.c
  *
@@ -238,11 +238,11 @@ asmlinkage int do_IRQ(unsigned long r4, unsigned long r5,
 	regs.syscall_nr = -1;	/* It's not system call */
 
 	/* Get IRQ number */
-	asm volatile("stc	r2_bank,%0\n\t"
+	asm volatile("stc	$r2_bank, %0\n\t"
 		     "shlr2	%0\n\t"
 		     "shlr2	%0\n\t"
 		     "shlr	%0\n\t"
-		     "add	#-16,%0\n\t"
+		     "add	#-16, %0\n\t"
 		     :"=z" (irq));
 
 	kstat.irqs[cpu][irq]++;
diff --git a/arch/sh/kernel/irq_imask.c b/arch/sh/kernel/irq_imask.c
index a3cf78b5f..380acf405 100644
--- a/arch/sh/kernel/irq_imask.c
+++ b/arch/sh/kernel/irq_imask.c
@@ -1,8 +1,8 @@
-/* $Id: irq_imask.c,v 1.2 2000/02/11 04:57:40 gniibe Exp $
+/* $Id: irq_imask.c,v 1.6 2000/03/06 14:11:32 gniibe Exp $
  *
  * linux/arch/sh/kernel/irq_imask.c
  *
- * Copyright (C) 1999  Niibe Yutaka
+ * Copyright (C) 1999, 2000  Niibe Yutaka
  *
  * Simple interrupt handling using IMASK of SR register.
  *
@@ -52,35 +52,40 @@ static struct hw_interrupt_type imask_irq_type = {
 	end_imask_irq
 };
 
-void disable_imask_irq(unsigned int irq)
+void static inline set_interrupt_registers(int ip)
 {
 	unsigned long __dummy;
 
+	asm volatile("ldc	%2, $r5_bank\n\t"
+		     "stc	$sr, %0\n\t"
+		     "and	#0xf0, %0\n\t"
+		     "shlr8	%0\n\t"
+		     "cmp/eq	#0x0f, %0\n\t"
+		     "bt	1f	! CLI-ed\n\t"
+		     "stc	$sr, %0\n\t"
+		     "and	%1, %0\n\t"
+		     "or	%2, %0\n\t"
+		     "ldc	%0, $sr\n"
+		     "1:"
+		     : "=&z" (__dummy)
+		     : "r" (~0xf0), "r" (ip << 4));
+}
+
+void disable_imask_irq(unsigned int irq)
+{
 	clear_bit(irq, &imask_mask);
 	if (interrupt_priority < IMASK_PRIORITY - irq)
 		interrupt_priority = IMASK_PRIORITY - irq;
 
-	asm volatile("stc	sr,%0\n\t"
-		     "and	%1,%0\n\t"
-		     "or	%2,%0\n\t"
-		     "ldc	%0,sr"
-		     : "=&r" (__dummy)
-		     : "r" (0xffffff0f), "r" (interrupt_priority << 4));
+	set_interrupt_registers(interrupt_priority);
 }
 
 static void enable_imask_irq(unsigned int irq)
 {
-	unsigned long __dummy;
-
 	set_bit(irq, &imask_mask);
 	interrupt_priority = IMASK_PRIORITY - ffz(imask_mask);
 
-	asm volatile("stc	sr,%0\n\t"
-		     "and	%1,%0\n\t"
-		     "or	%2,%0\n\t"
-		     "ldc	%0,sr"
-		     : "=&r" (__dummy)
-		     : "r" (0xffffff0f), "r" (interrupt_priority << 4));
+	set_interrupt_registers(interrupt_priority);
 }
 
 static void mask_and_ack_imask(unsigned int irq)
diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c
index 2ca91cb40..97cd1fe0c 100644
--- a/arch/sh/kernel/process.c
+++ b/arch/sh/kernel/process.c
@@ -1,4 +1,4 @@
-/* $Id: process.c,v 1.28 2000/03/05 02:16:15 gniibe Exp $
+/* $Id: process.c,v 1.33 2000/03/25 00:06:15 gniibe Exp $
  *
  *  linux/arch/sh/kernel/process.c
  *
@@ -94,7 +94,7 @@ void show_regs(struct pt_regs * regs)
 {
 	printk("\n");
 	printk("PC  : %08lx SP  : %08lx SR  : %08lx TEA : %08lx\n",
-	       regs->pc, regs->sp, regs->sr, ctrl_inl(MMU_TEA));
+	       regs->pc, regs->regs[15], regs->sr, ctrl_inl(MMU_TEA));
 	printk("R0  : %08lx R1  : %08lx R2  : %08lx R3  : %08lx\n",
 	       regs->regs[0],regs->regs[1],
 	       regs->regs[2],regs->regs[3]);
@@ -210,22 +210,22 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long usp,
 		struct task_struct *p, struct pt_regs *regs)
 {
 	struct pt_regs *childregs;
+#if defined(__SH4__)
 	struct task_struct *tsk = current;
 
-	childregs = ((struct pt_regs *)(THREAD_SIZE + (unsigned long) p)) - 1;
-	struct_cpy(childregs, regs);
-
-#if defined(__SH4__)
 	if (tsk != &init_task) {
 		unlazy_fpu(tsk);
-		struct_cpy(&p->thread.fpu, &current->thread.fpu);
+		p->thread.fpu = current->thread.fpu;
 		p->used_math = tsk->used_math;
 	}
 #endif
+	childregs = ((struct pt_regs *)(THREAD_SIZE + (unsigned long) p)) - 1;
+	*childregs = *regs;
+
 	if (user_mode(regs)) {
-		childregs->sp = usp;
+		childregs->regs[15] = usp;
 	} else {
-		childregs->sp = (unsigned long)p+2*PAGE_SIZE;
+		childregs->regs[15] = (unsigned long)p+2*PAGE_SIZE;
 	}
 	childregs->regs[0] = 0; /* Set return value for child */
 	childregs->sr |= SR_FD; /* Invalidate FPU flag */
@@ -244,7 +244,7 @@ void dump_thread(struct pt_regs * regs, struct user * dump)
 	dump->magic = CMAGIC;
 	dump->start_code = current->mm->start_code;
 	dump->start_data  = current->mm->start_data;
-	dump->start_stack = regs->sp & ~(PAGE_SIZE - 1);
+	dump->start_stack = regs->regs[15] & ~(PAGE_SIZE - 1);
 	dump->u_tsize = (current->mm->end_code - dump->start_code) >> PAGE_SHIFT;
 	dump->u_dsize = (current->mm->brk + (PAGE_SIZE-1) - dump->start_data) >> PAGE_SHIFT;
 	dump->u_ssize = (current->mm->start_stack - dump->start_stack +
@@ -279,7 +279,7 @@ asmlinkage int sys_fork(unsigned long r4, unsigned long r5,
 			unsigned long r6, unsigned long r7,
 			struct pt_regs regs)
 {
-	return do_fork(SIGCHLD, regs.sp, &regs);
+	return do_fork(SIGCHLD, regs.regs[15], &regs);
 }
 
 asmlinkage int sys_clone(unsigned long clone_flags, unsigned long newsp,
@@ -287,7 +287,7 @@ asmlinkage int sys_clone(unsigned long clone_flags, unsigned long newsp,
 			 struct pt_regs regs)
 {
 	if (!newsp)
-		newsp = regs.sp;
+		newsp = regs.regs[15];
 	return do_fork(clone_flags, newsp, &regs);
 }
 
@@ -305,7 +305,7 @@ asmlinkage int sys_vfork(unsigned long r4, unsigned long r5,
 			 unsigned long r6, unsigned long r7,
 			 struct pt_regs regs)
 {
-	return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs.sp, &regs);
+	return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs.regs[15], &regs);
 }
 
 /*
@@ -370,3 +370,12 @@ asmlinkage void print_syscall(int x)
 	       (init_task.flags&PF_USEDFPU)?'K':' ', (sr&SR_FD)?' ':'F');
 	restore_flags(flags);
 }
+
+asmlinkage void break_point_trap(void)
+{
+	/* Clear traicng.  */
+	ctrl_outw(0, UBC_BBRA);
+	ctrl_outw(0, UBC_BBRB);
+
+	force_sig(SIGTRAP, current);
+}
diff --git a/arch/sh/kernel/ptrace.c b/arch/sh/kernel/ptrace.c
index 2d69b5b7c..d4a1556b9 100644
--- a/arch/sh/kernel/ptrace.c
+++ b/arch/sh/kernel/ptrace.c
@@ -1,6 +1,13 @@
-/*
- * Surely this doesn't work... (we need to design ptrace for SupreH)
+/* $Id: ptrace.c,v 1.4 2000/03/22 13:59:01 gniibe Exp $
+ *
  * linux/arch/sh/kernel/ptrace.c
+ *
+ * Original x86 implementation:
+ *	By Ross Biro 1/23/92
+ *	edited by Linus Torvalds
+ *
+ * SuperH version:   Copyright (C) 1999, 2000  Kaz Kojima & Niibe Yutaka
+ *
  */
 
 #include <linux/kernel.h>
@@ -12,135 +19,126 @@
 #include <linux/ptrace.h>
 #include <linux/user.h>
 
+#include <asm/io.h>
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
 #include <asm/system.h>
 #include <asm/processor.h>
+#include <asm/mmu_context.h>
 
 /*
  * does not yet catch signals sent when the child dies.
  * in exit.c or in signal.c.
  */
 
-/* determines which flags the user has access to. */
-/* 1 = access 0 = no access */
-#define FLAG_MASK 0x00044dd5
-
-/* set's the trap flag. */
-#define TRAP_FLAG 0x100
-
 /*
- * Offset of eflags on child stack..
+ * This routine will get a word off of the process kernel stack.
  */
-#define EFL_OFFSET ((EFL-2)*4-sizeof(struct pt_regs))
-
-/*
- * this routine will get a word off of the processes privileged stack. 
- * the offset is how far from the base addr as stored in the TSS.  
- * this routine assumes that all the privileged stacks are in our
- * data space.
- */   
 static inline int get_stack_long(struct task_struct *task, int offset)
 {
 	unsigned char *stack;
 
-	stack = (unsigned char *)task->thread.sp;
+	stack = (unsigned char *)task + THREAD_SIZE - sizeof(struct pt_regs);
 	stack += offset;
 	return (*((int *)stack));
 }
 
 /*
- * this routine will put a word on the processes privileged stack. 
- * the offset is how far from the base addr as stored in the TSS.  
- * this routine assumes that all the privileged stacks are in our
- * data space.
+ * This routine will put a word on the process kernel stack.
  */
 static inline int put_stack_long(struct task_struct *task, int offset,
-	unsigned long data)
+				 unsigned long data)
 {
-	unsigned char * stack;
+	unsigned char *stack;
 
-	stack = (unsigned char *) task->thread.sp;
+	stack = (unsigned char *)task + THREAD_SIZE - sizeof(struct pt_regs);
 	stack += offset;
 	*(unsigned long *) stack = data;
 	return 0;
 }
 
-static int putreg(struct task_struct *child,
-	unsigned long regno, unsigned long value)
+static void
+compute_next_pc(struct pt_regs *regs, unsigned short inst,
+		unsigned long *pc1, unsigned long *pc2)
 {
-#if 0
-	switch (regno >> 2) {
-		case ORIG_EAX:
-			return -EIO;
-		case FS:
-			if (value && (value & 3) != 3)
-				return -EIO;
-			child->thread.fs = value;
-			return 0;
-		case GS:
-			if (value && (value & 3) != 3)
-				return -EIO;
-			child->thread.gs = value;
-			return 0;
-		case DS:
-		case ES:
-			if (value && (value & 3) != 3)
-				return -EIO;
-			value &= 0xffff;
-			break;
-		case SS:
-		case CS:
-			if ((value & 3) != 3)
-				return -EIO;
-			value &= 0xffff;
-			break;
-		case EFL:
-			value &= FLAG_MASK;
-			value |= get_stack_long(child, EFL_OFFSET) & ~FLAG_MASK;
+	int nib[4]
+		= { (inst >> 12) & 0xf,
+		    (inst >> 8) & 0xf,
+		    (inst >> 4) & 0xf,
+		    inst & 0xf};
+
+	/* bra & bsr */
+	if (nib[0] == 0xa || nib[0] == 0xb) {
+		*pc1 = regs->pc + 4 + ((short) ((inst & 0xfff) << 4) >> 3);
+		*pc2 = (unsigned long) -1;
+		return;
 	}
-	if (regno > GS*4)
-		regno -= 2*4;
-	put_stack_long(child, regno - sizeof(struct pt_regs), value);
-#endif
-	return 0;
+
+	/* bt & bf */
+	if (nib[0] == 0x8 && (nib[1] == 0x9 || nib[1] == 0xb)) {
+		*pc1 = regs->pc + 4 + ((char) (inst & 0xff) << 1);
+		*pc2 = regs->pc + 2;
+		return;
+	}
+
+	/* bt/s & bf/s */
+	if (nib[0] == 0x8 && (nib[1] == 0xd || nib[1] == 0xf)) {
+		*pc1 = regs->pc + 4 + ((char) (inst & 0xff) << 1);
+		*pc2 = regs->pc + 4;
+		return;
+	}
+
+	/* jmp & jsr */
+	if (nib[0] == 0x4 && nib[3] == 0xb
+	    && (nib[2] == 0x0 || nib[2] == 0x2)) {
+		*pc1 = regs->regs[nib[1]];
+		*pc2 = (unsigned long) -1;
+		return;
+	}
+
+	/* braf & bsrf */
+	if (nib[0] == 0x0 && nib[3] == 0x3
+	    && (nib[2] == 0x0 || nib[2] == 0x2)) {
+		*pc1 = regs->pc + 4 + regs->regs[nib[1]];
+		*pc2 = (unsigned long) -1;
+		return;
+	}
+
+	if (inst == 0x000b) {
+		*pc1 = regs->pr;
+		*pc2 = (unsigned long) -1;
+		return;
+	}
+
+	*pc1 = regs->pc + 2;
+	*pc2 = (unsigned long) -1;
+	return;
 }
 
-static unsigned long getreg(struct task_struct *child,
-	unsigned long regno)
+/* Tracing by user break controller.  */
+static void
+ubc_set_tracing(int asid, unsigned long nextpc1, unsigned nextpc2)
 {
-	unsigned long retval = ~0UL;
-
-#if 0
-	switch (regno >> 2) {
-		case FS:
-			retval = child->thread.fs;
-			break;
-		case GS:
-			retval = child->thread.gs;
-			break;
-		case DS:
-		case ES:
-		case SS:
-		case CS:
-			retval = 0xffff;
-			/* fall through */
-		default:
-			if (regno > GS*4)
-				regno -= 2*4;
-			regno = regno - sizeof(struct pt_regs);
-			retval &= get_stack_long(child, regno);
+	ctrl_outl(nextpc1, UBC_BARA);
+	ctrl_outb(asid, UBC_BASRA);
+	ctrl_outb(BAMR_12, UBC_BAMRA);
+	ctrl_outw(BBR_INST | BBR_READ, UBC_BBRA);
+
+	if (nextpc2 != (unsigned long) -1) {
+		ctrl_outl(nextpc2, UBC_BARB);
+		ctrl_outb(asid, UBC_BASRB);
+		ctrl_outb(BAMR_12, UBC_BAMRB);
+		ctrl_outw(BBR_INST | BBR_READ, UBC_BBRB);
 	}
-#endif
-	return retval;
+	ctrl_outw(BRCR_PCBA | BRCR_PCBB, UBC_BRCR);
 }
 
 asmlinkage int sys_ptrace(long request, long pid, long addr, long data)
 {
-	struct task_struct *child;
+	struct task_struct *child, *tsk = current;
 	struct user * dummy = NULL;
 	unsigned long flags;
-	int i, ret;
+	int ret;
 
 	lock_kernel();
 	ret = -EPERM;
@@ -163,16 +161,16 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data)
 	if (pid == 1)		/* you may not mess with init */
 		goto out;
 	if (request == PTRACE_ATTACH) {
-		if (child == current)
+		if (child == tsk)
 			goto out;
 		if ((!child->dumpable ||
-		    (current->uid != child->euid) ||
-		    (current->uid != child->suid) ||
-		    (current->uid != child->uid) ||
-	 	    (current->gid != child->egid) ||
-	 	    (current->gid != child->sgid) ||
-	 	    (!cap_issubset(child->cap_permitted, current->cap_permitted)) ||
-	 	    (current->gid != child->gid)) && !capable(CAP_SYS_PTRACE))
+		    (tsk->uid != child->euid) ||
+		    (tsk->uid != child->suid) ||
+		    (tsk->uid != child->uid) ||
+	 	    (tsk->gid != child->egid) ||
+	 	    (tsk->gid != child->sgid) ||
+	 	    (!cap_issubset(child->cap_permitted, tsk->cap_permitted)) ||
+	 	    (tsk->gid != child->gid)) && !capable(CAP_SYS_PTRACE))
 			goto out;
 		/* the same process cannot be attached many times */
 		if (child->flags & PF_PTRACED)
@@ -180,9 +178,9 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data)
 		child->flags |= PF_PTRACED;
 
 		write_lock_irqsave(&tasklist_lock, flags);
-		if (child->p_pptr != current) {
+		if (child->p_pptr != tsk) {
 			REMOVE_LINKS(child);
-			child->p_pptr = current;
+			child->p_pptr = tsk;
 			SET_LINKS(child);
 		}
 		write_unlock_irqrestore(&tasklist_lock, flags);
@@ -198,257 +196,180 @@ asmlinkage int sys_ptrace(long request, long pid, long addr, long data)
 		if (request != PTRACE_KILL)
 			goto out;
 	}
-	if (child->p_pptr != current)
+	if (child->p_pptr != tsk)
 		goto out;
 
 	switch (request) {
 	/* when I and D space are separate, these will need to be fixed. */
-		case PTRACE_PEEKTEXT: /* read word at location addr. */ 
-		case PTRACE_PEEKDATA: {
-			unsigned long tmp;
-			int copied;
-
-			copied = access_process_vm(child, addr, &tmp, sizeof(tmp), 0);
-			ret = -EIO;
-			if (copied != sizeof(tmp))
-				goto out;
-			ret = put_user(tmp,(unsigned long *) data);
-			goto out;
-		}
+	case PTRACE_PEEKTEXT: /* read word at location addr. */ 
+	case PTRACE_PEEKDATA: {
+		unsigned long tmp;
+		int copied;
+
+		copied = access_process_vm(child, addr, &tmp, sizeof(tmp), 0);
+		ret = -EIO;
+		if (copied != sizeof(tmp))
+			break;
+		ret = put_user(tmp,(unsigned long *) data);
+		break;
+	}
 
 	/* read the word at location addr in the USER area. */
-		case PTRACE_PEEKUSR: {
-			unsigned long tmp;
-
-			ret = -EIO;
-			if ((addr & 3) || addr < 0 || 
-			    addr > sizeof(struct user) - 3)
-				goto out;
-
-			tmp = 0;  /* Default return condition */
-			if(addr < 17*sizeof(long))
-				tmp = getreg(child, addr);
-#if 0
-			if(addr >= (long) &dummy->u_debugreg[0] &&
-			   addr <= (long) &dummy->u_debugreg[7]){
-				addr -= (long) &dummy->u_debugreg[0];
-				addr = addr >> 2;
-				tmp = child->thread.debugreg[addr];
-			};
-#endif
-			ret = put_user(tmp,(unsigned long *) data);
-			goto out;
-		}
+	case PTRACE_PEEKUSR: {
+		unsigned long tmp;
 
-      /* when I and D space are separate, this will have to be fixed. */
-		case PTRACE_POKETEXT: /* write the word at location addr. */
-		case PTRACE_POKEDATA:
-			ret = 0;
-			if (access_process_vm(child, addr, &data, sizeof(data), 1) == sizeof(data))
-				goto out;
-			ret = -EIO;
-			goto out;
+		ret = -EIO;
+		if ((addr & 3) || addr < 0 || 
+		    addr > sizeof(struct user) - 3)
+			break;
 
-		case PTRACE_POKEUSR: /* write the word at location addr in the USER area */
-			ret = -EIO;
-			if ((addr & 3) || addr < 0 || 
-			    addr > sizeof(struct user) - 3)
-				goto out;
-
-			if (addr < 17*sizeof(long)) {
-				ret = putreg(child, addr, data);
-				goto out;
-			}
-
-		  /* We need to be very careful here.  We implicitly
-		     want to modify a portion of the task_struct, and we
-		     have to be selective about what portions we allow someone
-		     to modify. */
-#if 0
-		  if(addr >= (long) &dummy->u_debugreg[0] &&
-		     addr <= (long) &dummy->u_debugreg[7]){
-
-			  if(addr == (long) &dummy->u_debugreg[4]) return -EIO;
-			  if(addr == (long) &dummy->u_debugreg[5]) return -EIO;
-			  if(addr < (long) &dummy->u_debugreg[4] &&
-			     ((unsigned long) data) >= TASK_SIZE-3) return -EIO;
-			  
-			  ret = -EIO;
-			  if(addr == (long) &dummy->u_debugreg[7]) {
-				  data &= ~DR_CONTROL_RESERVED;
-				  for(i=0; i<4; i++)
-					  if ((0x5f54 >> ((data >> (16 + 4*i)) & 0xf)) & 1)
-						  goto out;
-			  };
-
-			  addr -= (long) &dummy->u_debugreg;
-			  addr = addr >> 2;
-			  child->thread.debugreg[addr] = data;
-			  ret = 0;
-			  goto out;
-		  };
-#endif
-		  ret = -EIO;
-		  goto out;
-
-		case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */
-		case PTRACE_CONT: { /* restart after signal. */
-			long tmp;
-
-			ret = -EIO;
-			if ((unsigned long) data > _NSIG)
-				goto out;
-			if (request == PTRACE_SYSCALL)
-				child->flags |= PF_TRACESYS;
-			else
-				child->flags &= ~PF_TRACESYS;
-			child->exit_code = data;
-	/* make sure the single step bit is not set. */
-#if 0
-			tmp = get_stack_long(child, EFL_OFFSET) & ~TRAP_FLAG;
-			put_stack_long(child, EFL_OFFSET,tmp);
-#endif
-			wake_up_process(child);
+		if (addr < sizeof(struct pt_regs))
+			tmp = get_stack_long(child, addr);
+		else if (addr >= (long) &dummy->fpu &&
+			 addr < (long) &dummy->u_fpvalid) {
+			if (!child->used_math) {
+				if (addr == (long)&dummy->fpu.fpscr)
+					tmp = FPSCR_INIT;
+				else
+					tmp = 0;
+			} else
+				tmp = ((long *)&child->thread.fpu)
+					[(addr - (long)&dummy->fpu) >> 2];
+		} else if (addr == (long) &dummy->u_fpvalid)
+			tmp = child->used_math;
+		else
+			tmp = 0;
+		ret = put_user(tmp, (unsigned long *)data);
+		break;
+	}
+
+	/* when I and D space are separate, this will have to be fixed. */
+	case PTRACE_POKETEXT: /* write the word at location addr. */
+	case PTRACE_POKEDATA:
+		ret = 0;
+		if (access_process_vm(child, addr, &data, sizeof(data), 1) == sizeof(data))
+			break;
+		ret = -EIO;
+		break;
+
+	case PTRACE_POKEUSR: /* write the word at location addr in the USER area */
+		ret = -EIO;
+		if ((addr & 3) || addr < 0 || 
+		    addr > sizeof(struct user) - 3)
+			break;
+
+		if (addr < sizeof(struct pt_regs))
+			ret = put_stack_long(child, addr, data);
+		else if (addr >= (long) &dummy->fpu &&
+			 addr < (long) &dummy->u_fpvalid) {
+			child->used_math = 1;
+			((long *)&child->thread.fpu)
+				[(addr - (long)&dummy->fpu) >> 2] = data;
+			ret = 0;
+		} else if (addr == (long) &dummy->u_fpvalid) {
+			child->used_math = data?1:0;
 			ret = 0;
-			goto out;
 		}
+		break;
+
+	case PTRACE_SYSCALL: /* continue and stop at next (return from) syscall */
+	case PTRACE_CONT: { /* restart after signal. */
+		ret = -EIO;
+		if ((unsigned long) data > _NSIG)
+			break;
+		if (request == PTRACE_SYSCALL)
+			child->flags |= PF_TRACESYS;
+		else
+			child->flags &= ~PF_TRACESYS;
+		child->exit_code = data;
+		wake_up_process(child);
+		ret = 0;
+		break;
+	}
 
 /*
  * make the child exit.  Best I can do is send it a sigkill. 
  * perhaps it should be put in the status that it wants to 
  * exit.
  */
-		case PTRACE_KILL: {
-			long tmp;
+	case PTRACE_KILL: {
+		ret = 0;
+		if (child->state == TASK_ZOMBIE)	/* already dead */
+			break;
+		child->exit_code = SIGKILL;
+		wake_up_process(child);
+		break;
+	}
 
-			ret = 0;
-			if (child->state == TASK_ZOMBIE)	/* already dead */
-				goto out;
-			child->exit_code = SIGKILL;
-	/* make sure the single step bit is not set. */
-#if 0
-			tmp = get_stack_long(child, EFL_OFFSET) & ~TRAP_FLAG;
-			put_stack_long(child, EFL_OFFSET, tmp);
-#endif
-			wake_up_process(child);
-			goto out;
-		}
+	case PTRACE_SINGLESTEP: {  /* set the trap flag. */
+		long tmp, pc;
+		struct pt_regs *dummy = NULL;
+		struct pt_regs *regs;
+		unsigned long nextpc1, nextpc2;
+		unsigned short insn;
 
-		case PTRACE_SINGLESTEP: {  /* set the trap flag. */
-			long tmp;
+		ret = -EIO;
+		if ((unsigned long) data > _NSIG)
+			break;
+		child->flags &= ~PF_TRACESYS;
+		if ((child->flags & PF_DTRACE) == 0) {
+			/* Spurious delayed TF traps may occur */
+			child->flags |= PF_DTRACE;
+		}
 
-			ret = -EIO;
-			if ((unsigned long) data > _NSIG)
-				goto out;
-			child->flags &= ~PF_TRACESYS;
-			if ((child->flags & PF_DTRACE) == 0) {
-				/* Spurious delayed TF traps may occur */
-				child->flags |= PF_DTRACE;
-			}
-#if 0
-			tmp = get_stack_long(child, EFL_OFFSET) | TRAP_FLAG;
-			put_stack_long(child, EFL_OFFSET, tmp);
+		/* Compute next pc.  */
+		pc = get_stack_long(child, (long)&dummy->pc);
+		regs = (struct pt_regs *)((unsigned long)child + THREAD_SIZE - sizeof(struct pt_regs));
+		if (access_process_vm(child, pc&~3, &tmp, sizeof(tmp), 1) != sizeof(data))
+			break;
+ 
+#ifdef  __LITTLE_ENDIAN__
+		if (pc & 3)
+			insn = tmp >> 16;
+		else
+			insn = tmp & 0xffff;
+#else
+		if (pc & 3)
+			insn = tmp & 0xffff;
+		else
+			insn = tmp >> 16;
 #endif
-			child->exit_code = data;
-	/* give it a chance to run. */
-			wake_up_process(child);
-			ret = 0;
-			goto out;
-		}
+		compute_next_pc (regs, insn, &nextpc1, &nextpc2);
 
-		case PTRACE_DETACH: { /* detach a process that was attached. */
-			long tmp;
+		if (nextpc1 & 0x80000000)
+			break;
+		if (nextpc2 != (unsigned long) -1 && (nextpc2 & 0x80000000))
+			break;
 
-			ret = -EIO;
-			if ((unsigned long) data > _NSIG)
-				goto out;
-			child->flags &= ~(PF_PTRACED|PF_TRACESYS);
-			child->exit_code = data;
-			write_lock_irqsave(&tasklist_lock, flags);
-			REMOVE_LINKS(child);
-			child->p_pptr = child->p_opptr;
-			SET_LINKS(child);
-			write_unlock_irqrestore(&tasklist_lock, flags);
-	/* make sure the single step bit is not set. */
-#if 0
-			tmp = get_stack_long(child, EFL_OFFSET) & ~TRAP_FLAG;
-			put_stack_long(child, EFL_OFFSET, tmp);
-#endif
-			wake_up_process(child);
-			ret = 0;
-			goto out;
-		}
-#if 0
-		case PTRACE_GETREGS: { /* Get all gp regs from the child. */
-		  	if (!access_ok(VERIFY_WRITE, (unsigned *)data,
-				       17*sizeof(long)))
-			  {
-			    ret = -EIO;
-			    goto out;
-			  }
-			for ( i = 0; i < 17*sizeof(long); i += sizeof(long) )
-			  {
-			    __put_user(getreg(child, i),(unsigned long *) data);
-			    data += sizeof(long);
-			  }
-			ret = 0;
-			goto out;
-		  };
-
-		case PTRACE_SETREGS: { /* Set all gp regs in the child. */
-			unsigned long tmp;
-		  	if (!access_ok(VERIFY_READ, (unsigned *)data,
-				       17*sizeof(long)))
-			  {
-			    ret = -EIO;
-			    goto out;
-			  }
-			for ( i = 0; i < 17*sizeof(long); i += sizeof(long) )
-			  {
-			    __get_user(tmp, (unsigned long *) data);
-			    putreg(child, i, tmp);
-			    data += sizeof(long);
-			  }
-			ret = 0;
-			goto out;
-		  };
-
-		case PTRACE_GETFPREGS: { /* Get the child FPU state. */
-			if (!access_ok(VERIFY_WRITE, (unsigned *)data,
-				       sizeof(struct user_i387_struct)))
-			  {
-			    ret = -EIO;
-			    goto out;
-			  }
-			ret = 0;
-			if ( !child->used_math ) {
-			  /* Simulate an empty FPU. */
-			  child->thread.i387.hard.cwd = 0xffff037f;
-			  child->thread.i387.hard.swd = 0xffff0000;
-			  child->thread.i387.hard.twd = 0xffffffff;
-			}
-				__copy_to_user((void *)data, &child->thread.i387.hard,
-						sizeof(struct user_i387_struct));
-			goto out;
-		  };
-
-		case PTRACE_SETFPREGS: { /* Set the child FPU state. */
-			if (!access_ok(VERIFY_READ, (unsigned *)data,
-				       sizeof(struct user_i387_struct)))
-			  {
-			    ret = -EIO;
-			    goto out;
-			  }
-			child->used_math = 1;
-			  __copy_from_user(&child->thread.i387.hard, (void *)data,
-					   sizeof(struct user_i387_struct));
-			ret = 0;
-			goto out;
-		  };
-#endif
-		default:
-			ret = -EIO;
-			goto out;
+		ubc_set_tracing(child->mm->context & MMU_CONTEXT_ASID_MASK,
+				nextpc1, nextpc2);
+
+		child->exit_code = data;
+		/* give it a chance to run. */
+		wake_up_process(child);
+		ret = 0;
+		break;
+	}
+
+	case PTRACE_DETACH: { /* detach a process that was attached. */
+		ret = -EIO;
+		if ((unsigned long) data > _NSIG)
+			break;
+		child->flags &= ~(PF_PTRACED|PF_TRACESYS);
+		child->exit_code = data;
+		write_lock_irqsave(&tasklist_lock, flags);
+		REMOVE_LINKS(child);
+		child->p_pptr = child->p_opptr;
+		SET_LINKS(child);
+		write_unlock_irqrestore(&tasklist_lock, flags);
+		wake_up_process(child);
+		ret = 0;
+		break;
+	}
+
+	default:
+		ret = -EIO;
+		break;
 	}
 out:
 	unlock_kernel();
@@ -457,20 +378,22 @@ out:
 
 asmlinkage void syscall_trace(void)
 {
-	if ((current->flags & (PF_PTRACED|PF_TRACESYS))
-			!= (PF_PTRACED|PF_TRACESYS))
+	struct task_struct *tsk = current;
+
+	if ((tsk->flags & (PF_PTRACED|PF_TRACESYS))
+	    != (PF_PTRACED|PF_TRACESYS))
 		return;
-	current->exit_code = SIGTRAP;
-	current->state = TASK_STOPPED;
-	notify_parent(current, SIGCHLD);
+	tsk->exit_code = SIGTRAP;
+	tsk->state = TASK_STOPPED;
+	notify_parent(tsk, SIGCHLD);
 	schedule();
 	/*
 	 * this isn't the same as continuing with a signal, but it will do
 	 * for normal use.  strace only continues with a signal if the
 	 * stopping signal is not SIGTRAP.  -brl
 	 */
-	if (current->exit_code) {
-		send_sig(current->exit_code, current, 1);
-		current->exit_code = 0;
+	if (tsk->exit_code) {
+		send_sig(tsk->exit_code, tsk, 1);
+		tsk->exit_code = 0;
 	}
 }
diff --git a/arch/sh/kernel/signal.c b/arch/sh/kernel/signal.c
index 0c24acf73..4b11100ae 100644
--- a/arch/sh/kernel/signal.c
+++ b/arch/sh/kernel/signal.c
@@ -1,4 +1,4 @@
-/* $Id: signal.c,v 1.16 2000/01/29 11:31:31 gniibe Exp gniibe $
+/* $Id: signal.c,v 1.21 2000/03/11 14:06:21 gniibe Exp $
  *
  *  linux/arch/sh/kernel/signal.c
  *
@@ -6,7 +6,7 @@
  *
  *  1997-11-28  Modified for POSIX.1b signals by Richard Henderson
  *
- *  SuperH version:  Copyright (C) 1999  Niibe Yutaka
+ *  SuperH version:  Copyright (C) 1999, 2000  Niibe Yutaka & Kaz Kojima
  *
  */
 
@@ -125,7 +125,7 @@ sys_sigaltstack(const stack_t *uss, stack_t *uoss,
 		unsigned long r6, unsigned long r7,
 		struct pt_regs regs)
 {
-	return do_sigaltstack(uss, uoss, regs.sp);
+	return do_sigaltstack(uss, uoss, regs.regs[15]);
 }
 
 
@@ -136,7 +136,6 @@ sys_sigaltstack(const stack_t *uss, stack_t *uoss,
 struct sigframe
 {
 	struct sigcontext sc;
-	/* FPU data should come here: SH-3 has no FPU */
 	unsigned long extramask[_NSIG_WORDS-1];
 	char retcode[4];
 };
@@ -147,10 +146,38 @@ struct rt_sigframe
 	void *puc;
 	struct siginfo info;
 	struct ucontext uc;
-	/* FPU should come here: SH-3 has no FPU */
 	char retcode[4];
 };
 
+#if defined(__SH4__)
+static inline int restore_sigcontext_fpu(struct sigcontext *sc)
+{
+	current->used_math = 1;
+	return __copy_from_user(&tsk->thread.fpu.hard, &sc->sc_fpregs[0],
+				sizeof(long)*(NUM_FPU_REGS*2+2));
+}
+
+static inline int save_sigcontext_fpu(struct sigcontext *sc)
+{
+	struct task_struct *tsk = current;
+
+	if (!tsk->used_math) {
+		sc->owend_fp = 0;
+		return 0;
+	}
+
+	sc->owend_fp = 1;
+
+	/* This will cause a "finit" to be triggered by the next
+	   attempted FPU operation by the 'current' process.
+	   */
+	tsk->used_math = 0;
+
+	unlazy_fpu(tsk);
+	return __copy_to_user(&sc->sc_fpregs[0], &tsk->thread.fpu.hard,
+			      sizeof(long)*(NUM_FPU_REGS*2+2));
+}
+#endif
 
 static int
 restore_sigcontext(struct pt_regs *regs, struct sigcontext *sc, int *r0_p)
@@ -165,15 +192,28 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext *sc, int *r0_p)
 	COPY(regs[8]);	COPY(regs[9]);
 	COPY(regs[10]);	COPY(regs[11]);
 	COPY(regs[12]);	COPY(regs[13]);
-	COPY(regs[14]);	COPY(sp);
+	COPY(regs[14]);	COPY(regs[15]);
 	COPY(gbr);	COPY(mach);
 	COPY(macl);	COPY(pr);
 	COPY(sr);	COPY(pc);
 #undef COPY
 
+#if defined(__SH4__)
+	{
+		int owned_fp;
+		struct task_struct *tsk = current;
+
+		regs->sr |= SR_FD; /* Release FPU */
+		clear_fpu(tsk);
+		current->used_math = 0;
+		__get_user (owned_fp, &context->sc_ownedfp);
+		if (owned_fp)
+			err |= restore_sigcontext_fpu(sc);
+	}
+#endif
+
 	regs->syscall_nr = -1;		/* disable syscall checks */
 	err |= __get_user(*r0_p, &sc->sc_regs[0]);
-
 	return err;
 }
 
@@ -181,7 +221,7 @@ asmlinkage int sys_sigreturn(unsigned long r4, unsigned long r5,
 			     unsigned long r6, unsigned long r7,
 			     struct pt_regs regs)
 {
-	struct sigframe *frame = (struct sigframe *)regs.sp;
+	struct sigframe *frame = (struct sigframe *)regs.regs[15];
 	sigset_t set;
 	int r0;
 
@@ -214,7 +254,7 @@ asmlinkage int sys_rt_sigreturn(unsigned long r4, unsigned long r5,
 				unsigned long r6, unsigned long r7,
 				struct pt_regs regs)
 {
-	struct rt_sigframe *frame = (struct rt_sigframe *)regs.sp;
+	struct rt_sigframe *frame = (struct rt_sigframe *)regs.regs[15];
 	sigset_t set;
 	stack_t st;
 	int r0;
@@ -238,7 +278,7 @@ asmlinkage int sys_rt_sigreturn(unsigned long r4, unsigned long r5,
 		goto badframe;
 	/* It is more difficult to avoid calling this function than to
 	   call it and ignore errors.  */
-	do_sigaltstack(&st, NULL, regs.sp);
+	do_sigaltstack(&st, NULL, regs.regs[15]);
 
 	return r0;
 
@@ -265,12 +305,16 @@ setup_sigcontext(struct sigcontext *sc, struct pt_regs *regs,
 	COPY(regs[8]);	COPY(regs[9]);
 	COPY(regs[10]);	COPY(regs[11]);
 	COPY(regs[12]);	COPY(regs[13]);
-	COPY(regs[14]);	COPY(sp);
+	COPY(regs[14]);	COPY(regs[15]);
 	COPY(gbr);	COPY(mach);
 	COPY(macl);	COPY(pr);
 	COPY(sr);	COPY(pc);
 #undef COPY
 
+#if defined(__SH4__)
+	err |= save_sigcontext_fpu(sc);
+#endif
+
 	/* non-iBCS2 extensions.. */
 	err |= __put_user(mask, &sc->oldmask);
 
@@ -296,7 +340,7 @@ static void setup_frame(int sig, struct k_sigaction *ka,
 	int err = 0;
 	int signal;
 
-	frame = get_sigframe(ka, regs->sp, sizeof(*frame));
+	frame = get_sigframe(ka, regs->regs[15], sizeof(*frame));
 
 	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
 		goto give_sigsegv;
@@ -334,7 +378,7 @@ static void setup_frame(int sig, struct k_sigaction *ka,
 		goto give_sigsegv;
 
 	/* Set up registers for signal handler */
-	regs->sp = (unsigned long) frame;
+	regs->regs[15] = (unsigned long) frame;
 	regs->regs[4] = signal; /* Arg for signal handler */
 	regs->pc = (unsigned long) ka->sa.sa_handler;
 
@@ -361,7 +405,7 @@ static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	int err = 0;
 	int signal;
 
-	frame = get_sigframe(ka, regs->sp, sizeof(*frame));
+	frame = get_sigframe(ka, regs->regs[15], sizeof(*frame));
 
 	if (!access_ok(VERIFY_WRITE, frame, sizeof(*frame)))
 		goto give_sigsegv;
@@ -381,7 +425,8 @@ static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 	err |= __put_user(0, &frame->uc.uc_link);
 	err |= __put_user((void *)current->sas_ss_sp,
 			  &frame->uc.uc_stack.ss_sp);
-	err |= __put_user(sas_ss_flags(regs->sp), &frame->uc.uc_stack.ss_flags);
+	err |= __put_user(sas_ss_flags(regs->regs[15]),
+			  &frame->uc.uc_stack.ss_flags);
 	err |= __put_user(current->sas_ss_size, &frame->uc.uc_stack.ss_size);
 	err |= setup_sigcontext(&frame->uc.uc_mcontext,
 			        regs, set->sig[0]);
@@ -407,7 +452,7 @@ static void setup_rt_frame(int sig, struct k_sigaction *ka, siginfo_t *info,
 		goto give_sigsegv;
 
 	/* Set up registers for signal handler */
-	regs->sp = (unsigned long) frame;
+	regs->regs[15] = (unsigned long) frame;
 	regs->regs[4] = signal; /* Arg for signal handler */
 	regs->pc = (unsigned long) ka->sa.sa_handler;
 
diff --git a/arch/sh/lib/Makefile b/arch/sh/lib/Makefile
index 5f7d68755..7af24f1c3 100644
--- a/arch/sh/lib/Makefile
+++ b/arch/sh/lib/Makefile
@@ -3,7 +3,7 @@
 #
 
 .S.o:
-	$(CC) -D__ASSEMBLY__ $(AFLAGS) -traditional -c $< -o $*.o
+	$(CC) $(AFLAGS) -traditional -c $< -o $*.o
 
 L_TARGET = lib.a
 L_OBJS  = delay.o memcpy.o memset.o memmove.o memchr.o old-checksum.o \
diff --git a/arch/sh/mm/cache.c b/arch/sh/mm/cache.c
index f5c5200be..33bdf2114 100644
--- a/arch/sh/mm/cache.c
+++ b/arch/sh/mm/cache.c
@@ -1,4 +1,4 @@
-/* $Id: cache.c,v 1.9 2000/02/14 12:45:26 gniibe Exp $
+/* $Id: cache.c,v 1.10 2000/03/07 11:58:34 gniibe Exp $
  *
  *  linux/arch/sh/mm/cache.c
  *
@@ -67,36 +67,15 @@ static struct _cache_system_info cache_system_info;
 #define CACHE_IC_NUM_WAYS	  1
 #endif
 
-#define jump_to_p2(__dummy) \
-	asm volatile("mova	1f,%0\n\t" \
-		     "add	%1,%0\n\t" \
-		     "jmp	@r0		! Jump to P2 area\n\t" \
-		     " nop\n\t" \
-		     ".balign 4\n" \
-		     "1:" \
-		     : "=&z" (__dummy) \
-		     : "r" (0x20000000))
-
-#define back_to_p1(__dummy) \
-	asm volatile("nop;nop;nop;nop;nop;nop\n\t" \
-		     "mova	9f,%0\n\t" \
-		     "sub	%1,%0\n\t" \
-		     "jmp	@r0		! Back to P1 area\n\t" \
-		     " nop\n\t" \
-		     ".balign 4\n" \
-		     "9:" \
-		     : "=&z" (__dummy) \
-		     : "r" (0x20000000), "0" (__dummy))
-
 /* Write back caches to memory (if needed) and invalidates the caches */
 void cache_flush_area(unsigned long start, unsigned long end)
 {
-	unsigned long flags, __dummy;
+	unsigned long flags;
 	unsigned long addr, data, v, p;
 
 	start &= ~(L1_CACHE_BYTES-1);
 	save_and_cli(flags);
-	jump_to_p2(__dummy);
+	jump_to_P2();
 
 	for (v = start; v < end; v+=L1_CACHE_BYTES) {
 		p = __pa(v);
@@ -110,7 +89,7 @@ void cache_flush_area(unsigned long start, unsigned long end)
 			     : "m" (__m(v)));
 #endif
 	}
-	back_to_p1(__dummy);
+	back_to_P1();
 	restore_flags(flags);
 }
 
@@ -124,12 +103,12 @@ void cache_flush_area(unsigned long start, unsigned long end)
    the cache line. */
 void cache_purge_area(unsigned long start, unsigned long end)
 {
-	unsigned long flags, __dummy;
+	unsigned long flags;
 	unsigned long addr, data, v, p, j;
 
 	start &= ~(L1_CACHE_BYTES-1);
 	save_and_cli(flags);
-	jump_to_p2(__dummy);
+	jump_to_P2();
 
 	for (v = start; v < end; v+=L1_CACHE_BYTES) {
 		p = __pa(v);
@@ -150,19 +129,19 @@ void cache_purge_area(unsigned long start, unsigned long end)
 			     : "m" (__m(v)));
 #endif
 	}
-	back_to_p1(__dummy);
+	back_to_P1();
 	restore_flags(flags);
 }
 
 /* write back the dirty cache, but not invalidate the cache */
 void cache_wback_area(unsigned long start, unsigned long end)
 {
-	unsigned long flags, __dummy;
+	unsigned long flags;
 	unsigned long v;
 
 	start &= ~(L1_CACHE_BYTES-1);
 	save_and_cli(flags);
-	jump_to_p2(__dummy);
+	jump_to_P2();
 
 	for (v = start; v < end; v+=L1_CACHE_BYTES) {
 #if CACHE_IC_ADDRESS_ARRAY == CACHE_OC_ADDRESS_ARRAY
@@ -187,7 +166,7 @@ void cache_wback_area(unsigned long start, unsigned long end)
 			     : "m" (__m(v)));
 #endif
 	}
-	back_to_p1(__dummy);
+	back_to_P1();
 	restore_flags(flags);
 }
 
@@ -199,11 +178,11 @@ void cache_wback_area(unsigned long start, unsigned long end)
  */
 static void cache_wback_all(void)
 {
-	unsigned long flags, __dummy;
+	unsigned long flags;
 	unsigned long addr, data, i, j;
 
 	save_and_cli(flags);
-	jump_to_p2(__dummy);
+	jump_to_P2();
 
 	for (i=0; i<CACHE_OC_NUM_ENTRIES; i++) {
 		for (j=0; j<CACHE_OC_NUM_WAYS; j++) {
@@ -217,7 +196,7 @@ static void cache_wback_all(void)
 		}
 	}
 
-	back_to_p1(__dummy);
+	back_to_P1();
 	restore_flags(flags);
 }
 
@@ -225,9 +204,9 @@ static void
 detect_cpu_and_cache_system(void)
 {
 #if defined(__sh3__)
-	unsigned long __dummy, addr0, addr1, data0, data1, data2, data3;
+	unsigned long addr0, addr1, data0, data1, data2, data3;
 
-	jump_to_p2(__dummy);
+	jump_to_P2();
 	/* Check if the entry shadows or not.
 	 * When shadowed, it's 128-entry system.
 	 * Otherwise, it's 256-entry system.
@@ -245,7 +224,7 @@ detect_cpu_and_cache_system(void)
 	/* Invaliate them, in case the cache has been enabled already. */
 	ctrl_outl(data0&~0x00000001,addr0);
 	ctrl_outl(data2&~0x00000001,addr1);
-	back_to_p1(__dummy);
+	back_to_P1();
 
 	if (data0 == data1 && data2 == data3) {	/* Shadow */
 		cache_system_info.way_shift = 11;
@@ -265,7 +244,7 @@ detect_cpu_and_cache_system(void)
 
 void __init cache_init(void)
 {
-	unsigned long __dummy, ccr;
+	unsigned long ccr;
 
 	detect_cpu_and_cache_system();
 
@@ -277,19 +256,19 @@ void __init cache_init(void)
 		   we only need to flush the half of the caches. */
 		cache_wback_all();
 
-	jump_to_p2(__dummy);
+	jump_to_P2();
 	ctrl_outl(CCR_CACHE_INIT, CCR);
-	back_to_p1(__dummy);
+	back_to_P1();
 }
 
 #if defined(__SH4__)
 void flush_icache_page(struct vm_area_struct *vma, struct page *pg)
 {
-	unsigned long flags, __dummy;
+	unsigned long flags;
 	unsigned long addr, data, v;
 
 	save_and_cli(flags);
-	jump_to_p2(__dummy);
+	jump_to_P2();
 
 	v = page_address(pg);
 
@@ -303,18 +282,18 @@ void flush_icache_page(struct vm_area_struct *vma, struct page *pg)
 	data = (v&0xfffffc00); /* Valid=0 */
 	ctrl_outl(data,addr);
 
-	back_to_p1(__dummy);
+	back_to_P1();
 	restore_flags(flags);
 }
 
 void flush_icache_range(unsigned long start, unsigned long end)
 {
-	unsigned long flags, __dummy;
+	unsigned long flags;
 	unsigned long addr, data, v;
 
 	start &= ~(L1_CACHE_BYTES-1);
 	save_and_cli(flags);
-	jump_to_p2(__dummy);
+	jump_to_P2();
 
 	for (v = start; v < end; v+=L1_CACHE_BYTES) {
 		/* Write back O Cache */
@@ -327,22 +306,22 @@ void flush_icache_range(unsigned long start, unsigned long end)
 		data = (v&0xfffffc00); /* Valid=0 */
 		ctrl_outl(data,addr);
 	}
-	back_to_p1(__dummy);
+	back_to_P1();
 	restore_flags(flags);
 }
 
 void flush_cache_all(void)
 {
-	unsigned long flags,__dummy;
+	unsigned long flags;
 
 	/* Write back Operand Cache */
 	cache_wback_all ();
 
 	/* Then, invalidate Instruction Cache and Operand Cache */
 	save_and_cli(flags);
-	jump_to_p2(__dummy);
+	jump_to_P2();
 	ctrl_outl(CCR_CACHE_INIT, CCR);
-	back_to_p1(__dummy);
+	back_to_P1();
 	restore_flags(flags);
 }
 
@@ -356,12 +335,12 @@ void flush_cache_mm(struct mm_struct *mm)
 void flush_cache_range(struct mm_struct *mm, unsigned long start,
 		       unsigned long end)
 {
-	unsigned long flags, __dummy;
+	unsigned long flags;
 	unsigned long addr, data, v;
 
 	start &= ~(L1_CACHE_BYTES-1);
 	save_and_cli(flags);
-	jump_to_p2(__dummy);
+	jump_to_P2();
 
 	for (v = start; v < end; v+=L1_CACHE_BYTES) {
 		addr = CACHE_IC_ADDRESS_ARRAY |
@@ -372,7 +351,7 @@ void flush_cache_range(struct mm_struct *mm, unsigned long start,
 			(v&CACHE_OC_ENTRY_MASK) | 0x8 /* A-bit */;
 		ctrl_outl(data,addr);
 	}
-	back_to_p1(__dummy);
+	back_to_P1();
 	restore_flags(flags);
 }
 
diff --git a/arch/sh/mm/fault.c b/arch/sh/mm/fault.c
index 3b8e86e36..bf448bf8b 100644
--- a/arch/sh/mm/fault.c
+++ b/arch/sh/mm/fault.c
@@ -1,4 +1,4 @@
-/* $Id: fault.c,v 1.12 2000/03/01 11:15:27 gniibe Exp $
+/* $Id: fault.c,v 1.13 2000/03/07 12:05:24 gniibe Exp $
  *
  *  linux/arch/sh/mm/fault.c
  *  Copyright (C) 1999  Niibe Yutaka
@@ -283,9 +283,6 @@ static void __flush_tlb_page(struct mm_struct *mm, unsigned long page)
 {
 	unsigned long addr, data, asid;
 	unsigned long saved_asid = MMU_NO_ASID;
-#if defined(__SH4__)
-	int i;
-#endif
 
 	if (mm->context == NO_CONTEXT)
 		return;
@@ -305,20 +302,26 @@ static void __flush_tlb_page(struct mm_struct *mm, unsigned long page)
 	data = (page & 0xfffe0000) | asid; /* VALID bit is off */
 	ctrl_outl(data, addr);
 #elif defined(__SH4__)
+	jump_to_P2();
 	addr = MMU_UTLB_ADDRESS_ARRAY | MMU_PAGE_ASSOC_BIT;
 	data = page | asid; /* VALID bit is off */
 	ctrl_outl(data, addr);
-
-	for (i=0; i<4; i++) {
-		addr = MMU_ITLB_ADDRESS_ARRAY | (i<<8);
-		data = ctrl_inl(addr);
-		data &= ~0x300;
-		if (data == (page | asid)) {
-			ctrl_outl(data, addr);
-			break;
+#if 0 	/* Not need when using ASSOC. ??? */
+	{
+		int i;
+		for (i=0; i<4; i++) {
+			addr = MMU_ITLB_ADDRESS_ARRAY | (i<<8);
+			data = ctrl_inl(addr);
+			data &= ~0x300;
+			if (data == (page | asid)) {
+				ctrl_outl(data, addr);
+				break;
+			}
 		}
 	}
 #endif
+	back_to_P1();
+#endif
 	if (saved_asid != MMU_NO_ASID)
 		set_asid(saved_asid);
 }
diff --git a/arch/sh/mm/init.c b/arch/sh/mm/init.c
index 776389cac..931564cac 100644
--- a/arch/sh/mm/init.c
+++ b/arch/sh/mm/init.c
@@ -224,7 +224,7 @@ void __init paging_init(void)
 			zones_size[ZONE_DMA] = max_dma - start_pfn;
 			zones_size[ZONE_NORMAL] = low - max_dma;
 		}
-		free_area_init_node(0, 0, zones_size, __MEMORY_START);
+		free_area_init_node(0, 0, zones_size, __MEMORY_START, 0);
  	}
 }