sh: Fix up DSP context save/restore.

There were a number of issues with the DSP context save/restore code,
mostly left-over relics from when it was introduced on SH3-DSP with
little follow-up testing, resulting in things like task_pt_dspregs()
referencing incorrect state on the stack.

This follows the MIPS convention of tracking the DSP state in the
thread_struct and handling the state save/restore in switch_to() and
finish_arch_switch() respectively. The regset interface is also updated,
which allows us to finally be rid of task_pt_dspregs() and the special
cased task_pt_regs().

Signed-off-by: Michael Trimarchi <michael@evidence.eu.com>
Signed-off-by: Paul Mundt <lethal@linux-sh.org>

4 files changed, 25 insertions, 82 deletions

diff --git a/arch/sh/kernel/cpu/sh3/entry.S b/arch/sh/kernel/cpu/sh3/entry.S
index 55da0ff9848d..3cb531f233f2 100644
--- a/arch/sh/kernel/cpu/sh3/entry.S
+++ b/arch/sh/kernel/cpu/sh3/entry.S
@@ -254,40 +254,6 @@ restore_all:
 
         lds     k2, pr                  ! restore pr
         !
-#ifdef CONFIG_SH_DSP
-        mov.l   @r15+, k0               ! DSP mode marker
-        mov.l   5f, k1
-        cmp/eq  k0, k1                  ! Do we have a DSP stack frame?
-        bf      skip_restore
-
-        stc     sr, k0                  ! Enable CPU DSP mode
-        or      k1, k0                  ! (within kernel it may be disabled)
-        ldc     k0, sr
-        mov     r2, k0                  ! Backup r2
-
-        ! Restore DSP registers from stack
-        mov     r15, r2
-        movs.l  @r2+, a1
-        movs.l  @r2+, a0g
-        movs.l  @r2+, a1g
-        movs.l  @r2+, m0
-        movs.l  @r2+, m1
-        mov     r2, r15
-
-        lds.l   @r15+, a0
-        lds.l   @r15+, x0
-        lds.l   @r15+, x1
-        lds.l   @r15+, y0
-        lds.l   @r15+, y1
-        lds.l   @r15+, dsr
-        ldc.l   @r15+, rs
-        ldc.l   @r15+, re
-        ldc.l   @r15+, mod
-
-        mov     k0, r2                  ! Restore r2
-skip_restore:
-#endif
-        !
         ! Calculate new SR value
         mov     k3, k2                  ! original SR value
         mov     #0xf0, k1
@@ -358,7 +324,7 @@ general_exception:
         add     k0, k4
 0:
         ! Setup stack and save DSP context (k0 contains original r15 on return)
-        bsr     prepare_stack_save_dsp
+        bsr     prepare_stack
          nop
 
         ! Save registers / Switch to bank 0
@@ -374,15 +340,14 @@ general_exception:
 1:      .long   EXPEVT
 #endif
 
-! prepare_stack_save_dsp()
+! prepare_stack()
 ! - roll back gRB
 ! - switch to kernel stack
-! - save DSP
 ! k0 returns original sp (after roll back)
 ! k1 trashed
 ! k2 trashed
 
-prepare_stack_save_dsp:
+prepare_stack:
 #ifdef CONFIG_GUSA
         ! Check for roll back gRB (User and Kernel)
         mov     r15, k0
@@ -416,47 +381,9 @@ prepare_stack_save_dsp:
         mov     k1, r15         ! change to kernel stack
         !
 1:
-#ifdef CONFIG_SH_DSP
-        ! Save DSP context if needed
-        stc     sr, k1
-        mov     #0x10, k2
-        shll8   k2                      ! DSP=1 (0x00001000)
-        tst     k2, k1                  ! Check if in DSP mode (passed in k2)
-        bt/s    skip_save
-         mov    #0, k1                  ! Set marker for no stack frame
-
-        mov     k2, k1                  ! Save has-frame marker
-
-        ! Save DSP registers on stack
-        stc.l   mod, @-r15
-        stc.l   re, @-r15
-        stc.l   rs, @-r15
-        sts.l   dsr, @-r15
-        sts.l   y1, @-r15
-        sts.l   y0, @-r15
-        sts.l   x1, @-r15
-        sts.l   x0, @-r15
-        sts.l   a0, @-r15
-
-        ! GAS is broken, does not generate correct "movs.l Ds,@-As" instr.
-
-        ! FIXME: Make sure that this is still the case with newer toolchains,
-        ! as we're not at all interested in supporting ancient toolchains at
-        ! this point. -- PFM.
-
-        mov     r15, k2
-        .word   0xf653                  ! movs.l        a1, @-r2
-        .word   0xf6f3                  ! movs.l        a0g, @-r2
-        .word   0xf6d3                  ! movs.l        a1g, @-r2
-        .word   0xf6c3                  ! movs.l        m0, @-r2
-        .word   0xf6e3                  ! movs.l        m1, @-r2
-        mov     k2, r15
-
-skip_save:
-        mov.l   k1, @-r15               ! Push DSP mode marker onto stack
-#endif
         rts
          nop
+
 !
 ! 0x400: Instruction and Data TLB miss exception vector
 !
@@ -468,7 +395,7 @@ handle_exception:
         mova    exception_data, k0
 
         ! Setup stack and save DSP context (k0 contains original r15 on return)
-        bsr     prepare_stack_save_dsp
+        bsr     prepare_stack
          PREF(k0)
 
         ! Save registers / Switch to bank 0
@@ -572,7 +499,7 @@ ENTRY(handle_interrupt)
         mova    exception_data, k0
 
         ! Setup stack and save DSP context (k0 contains original r15 on return)
-        bsr     prepare_stack_save_dsp
+        bsr     prepare_stack
          PREF(k0)
 
         ! Save registers / Switch to bank 0
diff --git a/arch/sh/kernel/process_32.c b/arch/sh/kernel/process_32.c
index ddafbbbab2ab..0747fabd73a7 100644
--- a/arch/sh/kernel/process_32.c
+++ b/arch/sh/kernel/process_32.c
@@ -176,14 +176,26 @@ int copy_thread(int nr, unsigned long clone_flags, unsigned long usp,
 {
         struct thread_info *ti = task_thread_info(p);
         struct pt_regs *childregs;
-#if defined(CONFIG_SH_FPU)
+#if defined(CONFIG_SH_FPU) || defined(CONFIG_SH_DSP)
         struct task_struct *tsk = current;
+#endif
 
+#if defined(CONFIG_SH_FPU)
         unlazy_fpu(tsk, regs);
         p->thread.fpu = tsk->thread.fpu;
         copy_to_stopped_child_used_math(p);
 #endif
 
+#if defined(CONFIG_SH_DSP)
+        if (is_dsp_enabled(tsk)) {
+                /* We can use the __save_dsp or just copy the struct:
+                 * __save_dsp(p);
+                 * p->thread.dsp_status.status |= SR_DSP
+                 */
+                p->thread.dsp_status = tsk->thread.dsp_status;
+        }
+#endif
+
         childregs = task_pt_regs(p);
         *childregs = *regs;
 
diff --git a/arch/sh/kernel/ptrace_32.c b/arch/sh/kernel/ptrace_32.c
index 29ca09d24ef8..f7b22dd83b0c 100644
--- a/arch/sh/kernel/ptrace_32.c
+++ b/arch/sh/kernel/ptrace_32.c
@@ -200,7 +200,8 @@ static int dspregs_get(struct task_struct *target,
                        unsigned int pos, unsigned int count,
                        void *kbuf, void __user *ubuf)
 {
-        const struct pt_dspregs *regs = task_pt_dspregs(target);
+        const struct pt_dspregs *regs =
+                (struct pt_dspregs *)&target->thread.dsp_status.dsp_regs;
         int ret;
 
         ret = user_regset_copyout(&pos, &count, &kbuf, &ubuf, regs,
@@ -217,7 +218,8 @@ static int dspregs_set(struct task_struct *target,
                        unsigned int pos, unsigned int count,
                        const void *kbuf, const void __user *ubuf)
 {
-        struct pt_dspregs *regs = task_pt_dspregs(target);
+        struct pt_dspregs *regs =
+                (struct pt_dspregs *)&target->thread.dsp_status.dsp_regs;
         int ret;
 
         ret = user_regset_copyin(&pos, &count, &kbuf, &ubuf, regs,
diff --git a/arch/sh/kernel/traps_32.c b/arch/sh/kernel/traps_32.c
index 60dcf87ed019..30ca9c51e52d 100644
--- a/arch/sh/kernel/traps_32.c
+++ b/arch/sh/kernel/traps_32.c
@@ -664,6 +664,8 @@ asmlinkage void do_reserved_inst(unsigned long r4, unsigned long r5,
         if (is_dsp_inst(regs)) {
                 /* Enable DSP mode, and restart instruction. */
                 regs->sr |= SR_DSP;
+                /* Save DSP mode */
+                tsk->thread.dsp_status.status |= SR_DSP;
                 return;
         }
 #endif