diff mbox

[2.4] target-mips: add Config5.FRE support allowing Status.FR=0 emulation

Message ID 1426586167-1552-1-git-send-email-leon.alrae@imgtec.com
State New
Headers show

Commit Message

Leon Alrae March 17, 2015, 9:56 a.m. UTC
This relatively small architectural feature adds the following:

FIR.FREP: Read-only. If FREP=1, then Config5.FRE and Config5.UFE are available.

Config5.FRE: When enabled all single-precision FP arithmetic instructions,
             LWC1/LWXC1/MTC1, SWC1/SWXC1/MFC1 cause a Reserved Instructions
             exception.

Config5.UFE: Allows user to write/read Config5.FRE using CTC1/CFC1 instructions.

Enable the feature in MIPS64R6-generic CPU.

Signed-off-by: Leon Alrae <leon.alrae@imgtec.com>
---
 target-mips/cpu.h            |  13 +-
 target-mips/op_helper.c      |  34 +++++
 target-mips/translate.c      | 307 ++++++++++++++++++++++---------------------
 target-mips/translate_init.c |   9 +-
 4 files changed, 208 insertions(+), 155 deletions(-)

Comments

James Hogan March 17, 2015, 10:55 a.m. UTC | #1
Hi Leon,

On 17/03/15 09:56, Leon Alrae wrote:
> This relatively small architectural feature adds the following:
> 
> FIR.FREP: Read-only. If FREP=1, then Config5.FRE and Config5.UFE are available.
> 
> Config5.FRE: When enabled all single-precision FP arithmetic instructions,
>              LWC1/LWXC1/MTC1, SWC1/SWXC1/MFC1 cause a Reserved Instructions
>              exception.
> 
> Config5.UFE: Allows user to write/read Config5.FRE using CTC1/CFC1 instructions.
> 
> Enable the feature in MIPS64R6-generic CPU.
> 
> Signed-off-by: Leon Alrae <leon.alrae@imgtec.com>
> ---
>  target-mips/cpu.h            |  13 +-
>  target-mips/op_helper.c      |  34 +++++
>  target-mips/translate.c      | 307 ++++++++++++++++++++++---------------------
>  target-mips/translate_init.c |   9 +-
>  4 files changed, 208 insertions(+), 155 deletions(-)
> 
> diff --git a/target-mips/cpu.h b/target-mips/cpu.h
> index f9d2b4c..03eb888 100644
> --- a/target-mips/cpu.h
> +++ b/target-mips/cpu.h
> @@ -100,6 +100,7 @@ struct CPUMIPSFPUContext {
>      float_status fp_status;
>      /* fpu implementation/revision register (fir) */
>      uint32_t fcr0;
> +#define FCR0_FREP 29
>  #define FCR0_UFRP 28
>  #define FCR0_F64 22
>  #define FCR0_L 21
> @@ -462,6 +463,8 @@ struct CPUMIPSState {
>  #define CP0C5_CV         29
>  #define CP0C5_EVA        28
>  #define CP0C5_MSAEn      27
> +#define CP0C5_UFE        9
> +#define CP0C5_FRE        8
>  #define CP0C5_SBRI       6
>  #define CP0C5_UFR        2
>  #define CP0C5_NFExists   0
> @@ -514,7 +517,7 @@ struct CPUMIPSState {
>  #define EXCP_INST_NOTAVAIL 0x2 /* No valid instruction word for BadInstr */
>      uint32_t hflags;    /* CPU State */
>      /* TMASK defines different execution modes */
> -#define MIPS_HFLAG_TMASK  0x15807FF
> +#define MIPS_HFLAG_TMASK  0x35807FF
>  #define MIPS_HFLAG_MODE   0x00007 /* execution modes                    */
>      /* The KSU flags must be the lowest bits in hflags. The flag order
>         must be the same as defined for CP0 Status. This allows to use
> @@ -561,6 +564,7 @@ struct CPUMIPSState {
>  #define MIPS_HFLAG_SBRI  0x400000 /* R6 SDBBP causes RI excpt. in user mode */
>  #define MIPS_HFLAG_FBNSLOT 0x800000 /* Forbidden slot                   */
>  #define MIPS_HFLAG_MSA   0x1000000
> +#define MIPS_HFLAG_FRE   0x2000000 /* FRE enabled */
>      target_ulong btarget;        /* Jump / branch target               */
>      target_ulong bcond;          /* Branch condition (if needed)       */
>  
> @@ -843,7 +847,7 @@ static inline void compute_hflags(CPUMIPSState *env)
>      env->hflags &= ~(MIPS_HFLAG_COP1X | MIPS_HFLAG_64 | MIPS_HFLAG_CP0 |
>                       MIPS_HFLAG_F64 | MIPS_HFLAG_FPU | MIPS_HFLAG_KSU |
>                       MIPS_HFLAG_AWRAP | MIPS_HFLAG_DSP | MIPS_HFLAG_DSPR2 |
> -                     MIPS_HFLAG_SBRI | MIPS_HFLAG_MSA);
> +                     MIPS_HFLAG_SBRI | MIPS_HFLAG_MSA | MIPS_HFLAG_FRE);
>      if (!(env->CP0_Status & (1 << CP0St_EXL)) &&
>          !(env->CP0_Status & (1 << CP0St_ERL)) &&
>          !(env->hflags & MIPS_HFLAG_DM)) {
> @@ -924,6 +928,11 @@ static inline void compute_hflags(CPUMIPSState *env)
>              env->hflags |= MIPS_HFLAG_MSA;
>          }
>      }
> +    if (env->active_fpu.fcr0 & (1 << FCR0_FREP)) {
> +        if (env->CP0_Config5 & (1 << CP0C5_FRE)) {
> +            env->hflags |= MIPS_HFLAG_FRE;
> +        }
> +    }
>  }
>  
>  #ifndef CONFIG_USER_ONLY
> diff --git a/target-mips/op_helper.c b/target-mips/op_helper.c
> index 73a8e45..dd89068 100644
> --- a/target-mips/op_helper.c
> +++ b/target-mips/op_helper.c
> @@ -2303,6 +2303,16 @@ target_ulong helper_cfc1(CPUMIPSState *env, uint32_t reg)
>              }
>          }
>          break;
> +    case 5:
> +        /* FRE Support - read Config5.FRE bit */
> +        if (env->active_fpu.fcr0 & (1 << FCR0_FREP)) {
> +            if (env->CP0_Config5 & (1 << CP0C5_UFE)) {
> +                arg1 = (env->CP0_Config5 >> CP0C5_FRE) & 1;
> +            } else {
> +                helper_raise_exception(env, EXCP_RI);
> +            }
> +        }
> +        break;
>      case 25:
>          arg1 = ((env->active_fpu.fcr31 >> 24) & 0xfe) | ((env->active_fpu.fcr31 >> 23) & 0x1);
>          break;
> @@ -2347,6 +2357,30 @@ void helper_ctc1(CPUMIPSState *env, target_ulong arg1, uint32_t fs, uint32_t rt)
>              helper_raise_exception(env, EXCP_RI);
>          }
>          break;
> +    case 5:
> +        /* FRE Support - clear Config5.FRE bit */
> +        if (!((env->active_fpu.fcr0 & (1 << FCR0_FREP)) && (rt == 0))) {
> +            return;
> +        }

If rt != 0, is it really desired for a Config5 bit (which is privileged
state) to be modified? Assuming these behave similarly to UFR/UNFR, the
behaviour is architecturally UNPREDICTABLE when rt != $0, not UNDEFINED
(and at least UNFR is required to produce an RI exception in r5
implementations).

"UNPREDICTABLE operations must not read, write, or modify the contents
of memory or internal state which is inaccessible in the current
processor mode. For example, UNPREDICTABLE operations executed in user
mode must not access memory or internal state that is only accessible in
Kernel Mode or Debug Mode or in another process"

Probably same below and for UFR/UNFR really.

Should it be more like this?:
if (!((env->active_fpu.fcr0 & (1 << FCR0_FREP)) || (rt != 0))) {

That still ignores the potential RI that may or may not be required, but
that behaviour seems vaguely defined.

It also causes the UNPREDICTABLE rt != 0 case when FREP=1 to become a
no-op too which seems similarly safer, even though the FRE bit may
technically be "accessible" in user mode when FREP=1 && UFE=1, so not
impossible for an UNPREDICTABLE operation to clobber.


> +        if (env->CP0_Config5 & (1 << CP0C5_UFE)) {
> +            env->CP0_Config5 &= ~(1 << CP0C5_FRE);
> +            compute_hflags(env);
> +        } else {
> +            helper_raise_exception(env, EXCP_RI);
> +        }
> +        break;
> +    case 6:
> +        /* FRE Support - set Config5.FRE bit */
> +        if (!((env->active_fpu.fcr0 & (1 << FCR0_FREP)) && (rt == 0))) {
> +            return;
> +        }
> +        if (env->CP0_Config5 & (1 << CP0C5_UFE)) {
> +            env->CP0_Config5 |= (1 << CP0C5_FRE);
> +            compute_hflags(env);
> +        } else {
> +            helper_raise_exception(env, EXCP_RI);
> +        }
> +        break;
>      case 25:
>          if ((env->insn_flags & ISA_MIPS32R6) || (arg1 & 0xffffff00)) {
>              return;
> diff --git a/target-mips/translate.c b/target-mips/translate.c
> index 7030734..b9fcc8b 100644
> --- a/target-mips/translate.c
> +++ b/target-mips/translate.c
> @@ -1557,14 +1557,22 @@ static inline void gen_store_srsgpr (int from, int to)
>      }
>  }
>  
> +static inline void generate_exception(DisasContext *ctx, int excp);
> +

cleaner to just swap the "Floating point register moves" and "Tests"
groups of functions (as a separate commit)?

>  /* Floating point register moves. */
> -static void gen_load_fpr32(TCGv_i32 t, int reg)
> +static void gen_load_fpr32(DisasContext *ctx, TCGv_i32 t, int reg)
>  {
> +    if (ctx->hflags & MIPS_HFLAG_FRE) {
> +        generate_exception(ctx, EXCP_RI);

Maybe return to avoid generating dead code?

> +    }
>      tcg_gen_trunc_i64_i32(t, fpu_f64[reg]);
>  }
>  
> -static void gen_store_fpr32(TCGv_i32 t, int reg)
> +static void gen_store_fpr32(DisasContext *ctx, TCGv_i32 t, int reg)
>  {
> +    if (ctx->hflags & MIPS_HFLAG_FRE) {
> +        generate_exception(ctx, EXCP_RI);

same

> +    }
>      TCGv_i64 t64 = tcg_temp_new_i64();

declarations after code.

>      tcg_gen_extu_i32_i64(t64, t);
>      tcg_gen_deposit_i64(fpu_f64[reg], fpu_f64[reg], t64, 0, 32);

Rest looks okay AFAICT.

Thanks!
James
Leon Alrae March 17, 2015, 1:08 p.m. UTC | #2
Hi James,

On 17/03/2015 10:55, James Hogan wrote:
> Hi Leon,
> 
> On 17/03/15 09:56, Leon Alrae wrote:
>> This relatively small architectural feature adds the following:
>>
>> FIR.FREP: Read-only. If FREP=1, then Config5.FRE and Config5.UFE are available.
>>
>> Config5.FRE: When enabled all single-precision FP arithmetic instructions,
>>              LWC1/LWXC1/MTC1, SWC1/SWXC1/MFC1 cause a Reserved Instructions
>>              exception.
>>
>> Config5.UFE: Allows user to write/read Config5.FRE using CTC1/CFC1 instructions.
>>
>> Enable the feature in MIPS64R6-generic CPU.
>>
>> Signed-off-by: Leon Alrae <leon.alrae@imgtec.com>
>> ---
>>  target-mips/cpu.h            |  13 +-
>>  target-mips/op_helper.c      |  34 +++++
>>  target-mips/translate.c      | 307 ++++++++++++++++++++++---------------------
>>  target-mips/translate_init.c |   9 +-
>>  4 files changed, 208 insertions(+), 155 deletions(-)
>>
>> diff --git a/target-mips/cpu.h b/target-mips/cpu.h
>> index f9d2b4c..03eb888 100644
>> --- a/target-mips/cpu.h
>> +++ b/target-mips/cpu.h
>> @@ -100,6 +100,7 @@ struct CPUMIPSFPUContext {
>>      float_status fp_status;
>>      /* fpu implementation/revision register (fir) */
>>      uint32_t fcr0;
>> +#define FCR0_FREP 29
>>  #define FCR0_UFRP 28
>>  #define FCR0_F64 22
>>  #define FCR0_L 21
>> @@ -462,6 +463,8 @@ struct CPUMIPSState {
>>  #define CP0C5_CV         29
>>  #define CP0C5_EVA        28
>>  #define CP0C5_MSAEn      27
>> +#define CP0C5_UFE        9
>> +#define CP0C5_FRE        8
>>  #define CP0C5_SBRI       6
>>  #define CP0C5_UFR        2
>>  #define CP0C5_NFExists   0
>> @@ -514,7 +517,7 @@ struct CPUMIPSState {
>>  #define EXCP_INST_NOTAVAIL 0x2 /* No valid instruction word for BadInstr */
>>      uint32_t hflags;    /* CPU State */
>>      /* TMASK defines different execution modes */
>> -#define MIPS_HFLAG_TMASK  0x15807FF
>> +#define MIPS_HFLAG_TMASK  0x35807FF
>>  #define MIPS_HFLAG_MODE   0x00007 /* execution modes                    */
>>      /* The KSU flags must be the lowest bits in hflags. The flag order
>>         must be the same as defined for CP0 Status. This allows to use
>> @@ -561,6 +564,7 @@ struct CPUMIPSState {
>>  #define MIPS_HFLAG_SBRI  0x400000 /* R6 SDBBP causes RI excpt. in user mode */
>>  #define MIPS_HFLAG_FBNSLOT 0x800000 /* Forbidden slot                   */
>>  #define MIPS_HFLAG_MSA   0x1000000
>> +#define MIPS_HFLAG_FRE   0x2000000 /* FRE enabled */
>>      target_ulong btarget;        /* Jump / branch target               */
>>      target_ulong bcond;          /* Branch condition (if needed)       */
>>  
>> @@ -843,7 +847,7 @@ static inline void compute_hflags(CPUMIPSState *env)
>>      env->hflags &= ~(MIPS_HFLAG_COP1X | MIPS_HFLAG_64 | MIPS_HFLAG_CP0 |
>>                       MIPS_HFLAG_F64 | MIPS_HFLAG_FPU | MIPS_HFLAG_KSU |
>>                       MIPS_HFLAG_AWRAP | MIPS_HFLAG_DSP | MIPS_HFLAG_DSPR2 |
>> -                     MIPS_HFLAG_SBRI | MIPS_HFLAG_MSA);
>> +                     MIPS_HFLAG_SBRI | MIPS_HFLAG_MSA | MIPS_HFLAG_FRE);
>>      if (!(env->CP0_Status & (1 << CP0St_EXL)) &&
>>          !(env->CP0_Status & (1 << CP0St_ERL)) &&
>>          !(env->hflags & MIPS_HFLAG_DM)) {
>> @@ -924,6 +928,11 @@ static inline void compute_hflags(CPUMIPSState *env)
>>              env->hflags |= MIPS_HFLAG_MSA;
>>          }
>>      }
>> +    if (env->active_fpu.fcr0 & (1 << FCR0_FREP)) {
>> +        if (env->CP0_Config5 & (1 << CP0C5_FRE)) {
>> +            env->hflags |= MIPS_HFLAG_FRE;
>> +        }
>> +    }
>>  }
>>  
>>  #ifndef CONFIG_USER_ONLY
>> diff --git a/target-mips/op_helper.c b/target-mips/op_helper.c
>> index 73a8e45..dd89068 100644
>> --- a/target-mips/op_helper.c
>> +++ b/target-mips/op_helper.c
>> @@ -2303,6 +2303,16 @@ target_ulong helper_cfc1(CPUMIPSState *env, uint32_t reg)
>>              }
>>          }
>>          break;
>> +    case 5:
>> +        /* FRE Support - read Config5.FRE bit */
>> +        if (env->active_fpu.fcr0 & (1 << FCR0_FREP)) {
>> +            if (env->CP0_Config5 & (1 << CP0C5_UFE)) {
>> +                arg1 = (env->CP0_Config5 >> CP0C5_FRE) & 1;
>> +            } else {
>> +                helper_raise_exception(env, EXCP_RI);
>> +            }
>> +        }
>> +        break;
>>      case 25:
>>          arg1 = ((env->active_fpu.fcr31 >> 24) & 0xfe) | ((env->active_fpu.fcr31 >> 23) & 0x1);
>>          break;
>> @@ -2347,6 +2357,30 @@ void helper_ctc1(CPUMIPSState *env, target_ulong arg1, uint32_t fs, uint32_t rt)
>>              helper_raise_exception(env, EXCP_RI);
>>          }
>>          break;
>> +    case 5:
>> +        /* FRE Support - clear Config5.FRE bit */
>> +        if (!((env->active_fpu.fcr0 & (1 << FCR0_FREP)) && (rt == 0))) {
>> +            return;
>> +        }
> 
> If rt != 0, is it really desired for a Config5 bit (which is privileged
> state) to be modified? Assuming these behave similarly to UFR/UNFR, the
> behaviour is architecturally UNPREDICTABLE when rt != $0, not UNDEFINED
> (and at least UNFR is required to produce an RI exception in r5
> implementations).

Hmm, I believe the code above is correct and is doing exactly what you
described :) Note that "&& (rt == 0)" is inside parenthesis following
the logical NOT operator. It is no-op if rt != 0.

> 
> "UNPREDICTABLE operations must not read, write, or modify the contents
> of memory or internal state which is inaccessible in the current
> processor mode. For example, UNPREDICTABLE operations executed in user
> mode must not access memory or internal state that is only accessible in
> Kernel Mode or Debug Mode or in another process"
> 
> Probably same below and for UFR/UNFR really.
> 
> Should it be more like this?:
> if (!((env->active_fpu.fcr0 & (1 << FCR0_FREP)) || (rt != 0))) {
> 
> That still ignores the potential RI that may or may not be required, but
> that behaviour seems vaguely defined.
> 
> It also causes the UNPREDICTABLE rt != 0 case when FREP=1 to become a
> no-op too which seems similarly safer, even though the FRE bit may
> technically be "accessible" in user mode when FREP=1 && UFE=1, so not
> impossible for an UNPREDICTABLE operation to clobber.
> 
> 
>> +        if (env->CP0_Config5 & (1 << CP0C5_UFE)) {
>> +            env->CP0_Config5 &= ~(1 << CP0C5_FRE);
>> +            compute_hflags(env);
>> +        } else {
>> +            helper_raise_exception(env, EXCP_RI);
>> +        }
>> +        break;
>> +    case 6:
>> +        /* FRE Support - set Config5.FRE bit */
>> +        if (!((env->active_fpu.fcr0 & (1 << FCR0_FREP)) && (rt == 0))) {
>> +            return;
>> +        }
>> +        if (env->CP0_Config5 & (1 << CP0C5_UFE)) {
>> +            env->CP0_Config5 |= (1 << CP0C5_FRE);
>> +            compute_hflags(env);
>> +        } else {
>> +            helper_raise_exception(env, EXCP_RI);
>> +        }
>> +        break;
>>      case 25:
>>          if ((env->insn_flags & ISA_MIPS32R6) || (arg1 & 0xffffff00)) {
>>              return;
>> diff --git a/target-mips/translate.c b/target-mips/translate.c
>> index 7030734..b9fcc8b 100644
>> --- a/target-mips/translate.c
>> +++ b/target-mips/translate.c
>> @@ -1557,14 +1557,22 @@ static inline void gen_store_srsgpr (int from, int to)
>>      }
>>  }
>>  
>> +static inline void generate_exception(DisasContext *ctx, int excp);
>> +
> 
> cleaner to just swap the "Floating point register moves" and "Tests"
> groups of functions (as a separate commit)?

Indeed. I'll add a separate patch.

> 
>>  /* Floating point register moves. */
>> -static void gen_load_fpr32(TCGv_i32 t, int reg)
>> +static void gen_load_fpr32(DisasContext *ctx, TCGv_i32 t, int reg)
>>  {
>> +    if (ctx->hflags & MIPS_HFLAG_FRE) {
>> +        generate_exception(ctx, EXCP_RI);
> 
> Maybe return to avoid generating dead code?

The reason is to avoid leaving the TCG temp marked as TEMP_VAL_DEAD
which would cause assertion failures in TCG as we still generate code
using this temp after returning from gen_load_fpr().

> 
>> +    }
>>      tcg_gen_trunc_i64_i32(t, fpu_f64[reg]);
>>  }
>>  
>> -static void gen_store_fpr32(TCGv_i32 t, int reg)
>> +static void gen_store_fpr32(DisasContext *ctx, TCGv_i32 t, int reg)
>>  {
>> +    if (ctx->hflags & MIPS_HFLAG_FRE) {
>> +        generate_exception(ctx, EXCP_RI);
> 
> same
> 
>> +    }
>>      TCGv_i64 t64 = tcg_temp_new_i64();
> 
> declarations after code.

Thanks for spotting this!

> 
>>      tcg_gen_extu_i32_i64(t64, t);
>>      tcg_gen_deposit_i64(fpu_f64[reg], fpu_f64[reg], t64, 0, 32);
> 
> Rest looks okay AFAICT.
> 

Leon
James Hogan March 17, 2015, 3:08 p.m. UTC | #3
On 17/03/15 13:08, Leon Alrae wrote:
> Hi James,
> 
> On 17/03/2015 10:55, James Hogan wrote:
>> Hi Leon,
>>
>> On 17/03/15 09:56, Leon Alrae wrote:
>>> +    case 5:
>>> +        /* FRE Support - clear Config5.FRE bit */
>>> +        if (!((env->active_fpu.fcr0 & (1 << FCR0_FREP)) && (rt == 0))) {
>>> +            return;
>>> +        }
>>
>> If rt != 0, is it really desired for a Config5 bit (which is privileged
>> state) to be modified? Assuming these behave similarly to UFR/UNFR, the
>> behaviour is architecturally UNPREDICTABLE when rt != $0, not UNDEFINED
>> (and at least UNFR is required to produce an RI exception in r5
>> implementations).
> 
> Hmm, I believe the code above is correct and is doing exactly what you
> described :) Note that "&& (rt == 0)" is inside parenthesis following
> the logical NOT operator. It is no-op if rt != 0.

Yep, my apologies. Stupid brackets :-).

>>>  /* Floating point register moves. */
>>> -static void gen_load_fpr32(TCGv_i32 t, int reg)
>>> +static void gen_load_fpr32(DisasContext *ctx, TCGv_i32 t, int reg)
>>>  {
>>> +    if (ctx->hflags & MIPS_HFLAG_FRE) {
>>> +        generate_exception(ctx, EXCP_RI);
>>
>> Maybe return to avoid generating dead code?
> 
> The reason is to avoid leaving the TCG temp marked as TEMP_VAL_DEAD
> which would cause assertion failures in TCG as we still generate code
> using this temp after returning from gen_load_fpr().

Okay, fair enough.

Cheers
James
diff mbox

Patch

diff --git a/target-mips/cpu.h b/target-mips/cpu.h
index f9d2b4c..03eb888 100644
--- a/target-mips/cpu.h
+++ b/target-mips/cpu.h
@@ -100,6 +100,7 @@  struct CPUMIPSFPUContext {
     float_status fp_status;
     /* fpu implementation/revision register (fir) */
     uint32_t fcr0;
+#define FCR0_FREP 29
 #define FCR0_UFRP 28
 #define FCR0_F64 22
 #define FCR0_L 21
@@ -462,6 +463,8 @@  struct CPUMIPSState {
 #define CP0C5_CV         29
 #define CP0C5_EVA        28
 #define CP0C5_MSAEn      27
+#define CP0C5_UFE        9
+#define CP0C5_FRE        8
 #define CP0C5_SBRI       6
 #define CP0C5_UFR        2
 #define CP0C5_NFExists   0
@@ -514,7 +517,7 @@  struct CPUMIPSState {
 #define EXCP_INST_NOTAVAIL 0x2 /* No valid instruction word for BadInstr */
     uint32_t hflags;    /* CPU State */
     /* TMASK defines different execution modes */
-#define MIPS_HFLAG_TMASK  0x15807FF
+#define MIPS_HFLAG_TMASK  0x35807FF
 #define MIPS_HFLAG_MODE   0x00007 /* execution modes                    */
     /* The KSU flags must be the lowest bits in hflags. The flag order
        must be the same as defined for CP0 Status. This allows to use
@@ -561,6 +564,7 @@  struct CPUMIPSState {
 #define MIPS_HFLAG_SBRI  0x400000 /* R6 SDBBP causes RI excpt. in user mode */
 #define MIPS_HFLAG_FBNSLOT 0x800000 /* Forbidden slot                   */
 #define MIPS_HFLAG_MSA   0x1000000
+#define MIPS_HFLAG_FRE   0x2000000 /* FRE enabled */
     target_ulong btarget;        /* Jump / branch target               */
     target_ulong bcond;          /* Branch condition (if needed)       */
 
@@ -843,7 +847,7 @@  static inline void compute_hflags(CPUMIPSState *env)
     env->hflags &= ~(MIPS_HFLAG_COP1X | MIPS_HFLAG_64 | MIPS_HFLAG_CP0 |
                      MIPS_HFLAG_F64 | MIPS_HFLAG_FPU | MIPS_HFLAG_KSU |
                      MIPS_HFLAG_AWRAP | MIPS_HFLAG_DSP | MIPS_HFLAG_DSPR2 |
-                     MIPS_HFLAG_SBRI | MIPS_HFLAG_MSA);
+                     MIPS_HFLAG_SBRI | MIPS_HFLAG_MSA | MIPS_HFLAG_FRE);
     if (!(env->CP0_Status & (1 << CP0St_EXL)) &&
         !(env->CP0_Status & (1 << CP0St_ERL)) &&
         !(env->hflags & MIPS_HFLAG_DM)) {
@@ -924,6 +928,11 @@  static inline void compute_hflags(CPUMIPSState *env)
             env->hflags |= MIPS_HFLAG_MSA;
         }
     }
+    if (env->active_fpu.fcr0 & (1 << FCR0_FREP)) {
+        if (env->CP0_Config5 & (1 << CP0C5_FRE)) {
+            env->hflags |= MIPS_HFLAG_FRE;
+        }
+    }
 }
 
 #ifndef CONFIG_USER_ONLY
diff --git a/target-mips/op_helper.c b/target-mips/op_helper.c
index 73a8e45..dd89068 100644
--- a/target-mips/op_helper.c
+++ b/target-mips/op_helper.c
@@ -2303,6 +2303,16 @@  target_ulong helper_cfc1(CPUMIPSState *env, uint32_t reg)
             }
         }
         break;
+    case 5:
+        /* FRE Support - read Config5.FRE bit */
+        if (env->active_fpu.fcr0 & (1 << FCR0_FREP)) {
+            if (env->CP0_Config5 & (1 << CP0C5_UFE)) {
+                arg1 = (env->CP0_Config5 >> CP0C5_FRE) & 1;
+            } else {
+                helper_raise_exception(env, EXCP_RI);
+            }
+        }
+        break;
     case 25:
         arg1 = ((env->active_fpu.fcr31 >> 24) & 0xfe) | ((env->active_fpu.fcr31 >> 23) & 0x1);
         break;
@@ -2347,6 +2357,30 @@  void helper_ctc1(CPUMIPSState *env, target_ulong arg1, uint32_t fs, uint32_t rt)
             helper_raise_exception(env, EXCP_RI);
         }
         break;
+    case 5:
+        /* FRE Support - clear Config5.FRE bit */
+        if (!((env->active_fpu.fcr0 & (1 << FCR0_FREP)) && (rt == 0))) {
+            return;
+        }
+        if (env->CP0_Config5 & (1 << CP0C5_UFE)) {
+            env->CP0_Config5 &= ~(1 << CP0C5_FRE);
+            compute_hflags(env);
+        } else {
+            helper_raise_exception(env, EXCP_RI);
+        }
+        break;
+    case 6:
+        /* FRE Support - set Config5.FRE bit */
+        if (!((env->active_fpu.fcr0 & (1 << FCR0_FREP)) && (rt == 0))) {
+            return;
+        }
+        if (env->CP0_Config5 & (1 << CP0C5_UFE)) {
+            env->CP0_Config5 |= (1 << CP0C5_FRE);
+            compute_hflags(env);
+        } else {
+            helper_raise_exception(env, EXCP_RI);
+        }
+        break;
     case 25:
         if ((env->insn_flags & ISA_MIPS32R6) || (arg1 & 0xffffff00)) {
             return;
diff --git a/target-mips/translate.c b/target-mips/translate.c
index 7030734..b9fcc8b 100644
--- a/target-mips/translate.c
+++ b/target-mips/translate.c
@@ -1557,14 +1557,22 @@  static inline void gen_store_srsgpr (int from, int to)
     }
 }
 
+static inline void generate_exception(DisasContext *ctx, int excp);
+
 /* Floating point register moves. */
-static void gen_load_fpr32(TCGv_i32 t, int reg)
+static void gen_load_fpr32(DisasContext *ctx, TCGv_i32 t, int reg)
 {
+    if (ctx->hflags & MIPS_HFLAG_FRE) {
+        generate_exception(ctx, EXCP_RI);
+    }
     tcg_gen_trunc_i64_i32(t, fpu_f64[reg]);
 }
 
-static void gen_store_fpr32(TCGv_i32 t, int reg)
+static void gen_store_fpr32(DisasContext *ctx, TCGv_i32 t, int reg)
 {
+    if (ctx->hflags & MIPS_HFLAG_FRE) {
+        generate_exception(ctx, EXCP_RI);
+    }
     TCGv_i64 t64 = tcg_temp_new_i64();
     tcg_gen_extu_i32_i64(t64, t);
     tcg_gen_deposit_i64(fpu_f64[reg], fpu_f64[reg], t64, 0, 32);
@@ -1579,7 +1587,7 @@  static void gen_load_fpr32h(DisasContext *ctx, TCGv_i32 t, int reg)
         tcg_gen_trunc_i64_i32(t, t64);
         tcg_temp_free_i64(t64);
     } else {
-        gen_load_fpr32(t, reg | 1);
+        gen_load_fpr32(ctx, t, reg | 1);
     }
 }
 
@@ -1591,7 +1599,7 @@  static void gen_store_fpr32h(DisasContext *ctx, TCGv_i32 t, int reg)
         tcg_gen_deposit_i64(fpu_f64[reg], fpu_f64[reg], t64, 32, 32);
         tcg_temp_free_i64(t64);
     } else {
-        gen_store_fpr32(t, reg | 1);
+        gen_store_fpr32(ctx, t, reg | 1);
     }
 }
 
@@ -1819,7 +1827,7 @@  static inline void check_mips_64(DisasContext *ctx)
    calling interface for 32 and 64-bit FPRs.  No sense in changing
    all callers for gen_load_fpr32 when we need the CTX parameter for
    this one use.  */
-#define gen_ldcmp_fpr32(ctx, x, y) gen_load_fpr32(x, y)
+#define gen_ldcmp_fpr32(ctx, x, y) gen_load_fpr32(ctx, x, y)
 #define gen_ldcmp_fpr64(ctx, x, y) gen_load_fpr64(ctx, x, y)
 #define FOP_CONDS(type, abs, fmt, ifmt, bits)                                 \
 static inline void gen_cmp ## type ## _ ## fmt(DisasContext *ctx, int n,      \
@@ -1963,7 +1971,7 @@  static inline void gen_r6_cmp_ ## fmt(DisasContext * ctx, int n,        \
 }
 
 FOP_CONDNS(d, FMT_D, 64, gen_store_fpr64(ctx, fp0, fd))
-FOP_CONDNS(s, FMT_S, 32, gen_store_fpr32(fp0, fd))
+FOP_CONDNS(s, FMT_S, 32, gen_store_fpr32(ctx, fp0, fd))
 #undef FOP_CONDNS
 #undef gen_ldcmp_fpr32
 #undef gen_ldcmp_fpr64
@@ -2348,7 +2356,7 @@  static void gen_flt_ldst (DisasContext *ctx, uint32_t opc, int ft,
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
             tcg_gen_qemu_ld_i32(fp0, t0, ctx->mem_idx, MO_TESL);
-            gen_store_fpr32(fp0, ft);
+            gen_store_fpr32(ctx, fp0, ft);
             tcg_temp_free_i32(fp0);
         }
         opn = "lwc1";
@@ -2356,7 +2364,7 @@  static void gen_flt_ldst (DisasContext *ctx, uint32_t opc, int ft,
     case OPC_SWC1:
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
-            gen_load_fpr32(fp0, ft);
+            gen_load_fpr32(ctx, fp0, ft);
             tcg_gen_qemu_st_i32(fp0, t0, ctx->mem_idx, MO_TEUL);
             tcg_temp_free_i32(fp0);
         }
@@ -7557,7 +7565,7 @@  static void gen_mftr(CPUMIPSState *env, DisasContext *ctx, int rt, int rd,
         if (h == 0) {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, rt);
+            gen_load_fpr32(ctx, fp0, rt);
             tcg_gen_ext_i32_tl(t0, fp0);
             tcg_temp_free_i32(fp0);
         } else {
@@ -7756,7 +7764,7 @@  static void gen_mttr(CPUMIPSState *env, DisasContext *ctx, int rd, int rt,
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
             tcg_gen_trunc_tl_i32(fp0, t0);
-            gen_store_fpr32(fp0, rd);
+            gen_store_fpr32(ctx, fp0, rd);
             tcg_temp_free_i32(fp0);
         } else {
             TCGv_i32 fp0 = tcg_temp_new_i32();
@@ -8346,7 +8354,7 @@  static void gen_cp1 (DisasContext *ctx, uint32_t opc, int rt, int fs)
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             tcg_gen_ext_i32_tl(t0, fp0);
             tcg_temp_free_i32(fp0);
         }
@@ -8359,7 +8367,7 @@  static void gen_cp1 (DisasContext *ctx, uint32_t opc, int rt, int fs)
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
             tcg_gen_trunc_tl_i32(fp0, t0);
-            gen_store_fpr32(fp0, fs);
+            gen_store_fpr32(ctx, fp0, fs);
             tcg_temp_free_i32(fp0);
         }
         opn = "mtc1";
@@ -8457,7 +8465,8 @@  static void gen_movci (DisasContext *ctx, int rd, int rs, int cc, int tf)
     gen_set_label(l1);
 }
 
-static inline void gen_movcf_s (int fs, int fd, int cc, int tf)
+static inline void gen_movcf_s(DisasContext *ctx, int fs, int fd, int cc,
+                               int tf)
 {
     int cond;
     TCGv_i32 t0 = tcg_temp_new_i32();
@@ -8470,8 +8479,8 @@  static inline void gen_movcf_s (int fs, int fd, int cc, int tf)
 
     tcg_gen_andi_i32(t0, fpu_fcr31, 1 << get_fp_bit(cc));
     tcg_gen_brcondi_i32(cond, t0, 0, l1);
-    gen_load_fpr32(t0, fs);
-    gen_store_fpr32(t0, fd);
+    gen_load_fpr32(ctx, t0, fs);
+    gen_store_fpr32(ctx, t0, fd);
     gen_set_label(l1);
     tcg_temp_free_i32(t0);
 }
@@ -8513,8 +8522,8 @@  static inline void gen_movcf_ps(DisasContext *ctx, int fs, int fd,
 
     tcg_gen_andi_i32(t0, fpu_fcr31, 1 << get_fp_bit(cc));
     tcg_gen_brcondi_i32(cond, t0, 0, l1);
-    gen_load_fpr32(t0, fs);
-    gen_store_fpr32(t0, fd);
+    gen_load_fpr32(ctx, t0, fs);
+    gen_store_fpr32(ctx, t0, fd);
     gen_set_label(l1);
 
     tcg_gen_andi_i32(t0, fpu_fcr31, 1 << get_fp_bit(cc+1));
@@ -8532,9 +8541,9 @@  static void gen_sel_s(DisasContext *ctx, enum fopcode op1, int fd, int ft,
     TCGv_i32 fp0 = tcg_temp_new_i32();
     TCGv_i32 fp1 = tcg_temp_new_i32();
     TCGv_i32 fp2 = tcg_temp_new_i32();
-    gen_load_fpr32(fp0, fd);
-    gen_load_fpr32(fp1, ft);
-    gen_load_fpr32(fp2, fs);
+    gen_load_fpr32(ctx, fp0, fd);
+    gen_load_fpr32(ctx, fp1, ft);
+    gen_load_fpr32(ctx, fp2, fs);
 
     switch (op1) {
     case OPC_SEL_S:
@@ -8555,7 +8564,7 @@  static void gen_sel_s(DisasContext *ctx, enum fopcode op1, int fd, int ft,
         break;
     }
 
-    gen_store_fpr32(fp0, fd);
+    gen_store_fpr32(ctx, fp0, fd);
     tcg_temp_free_i32(fp2);
     tcg_temp_free_i32(fp1);
     tcg_temp_free_i32(fp0);
@@ -8648,11 +8657,11 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp0 = tcg_temp_new_i32();
             TCGv_i32 fp1 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
             gen_helper_float_add_s(fp0, cpu_env, fp0, fp1);
             tcg_temp_free_i32(fp1);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "add.s";
@@ -8663,11 +8672,11 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp0 = tcg_temp_new_i32();
             TCGv_i32 fp1 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
             gen_helper_float_sub_s(fp0, cpu_env, fp0, fp1);
             tcg_temp_free_i32(fp1);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "sub.s";
@@ -8678,11 +8687,11 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp0 = tcg_temp_new_i32();
             TCGv_i32 fp1 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
             gen_helper_float_mul_s(fp0, cpu_env, fp0, fp1);
             tcg_temp_free_i32(fp1);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "mul.s";
@@ -8693,11 +8702,11 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp0 = tcg_temp_new_i32();
             TCGv_i32 fp1 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
             gen_helper_float_div_s(fp0, cpu_env, fp0, fp1);
             tcg_temp_free_i32(fp1);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "div.s";
@@ -8707,9 +8716,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_helper_float_sqrt_s(fp0, cpu_env, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "sqrt.s";
@@ -8718,9 +8727,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_helper_float_abs_s(fp0, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "abs.s";
@@ -8729,8 +8738,8 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
-            gen_store_fpr32(fp0, fd);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "mov.s";
@@ -8739,9 +8748,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_helper_float_chs_s(fp0, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "neg.s";
@@ -8752,7 +8761,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp32 = tcg_temp_new_i32();
             TCGv_i64 fp64 = tcg_temp_new_i64();
 
-            gen_load_fpr32(fp32, fs);
+            gen_load_fpr32(ctx, fp32, fs);
             gen_helper_float_roundl_s(fp64, cpu_env, fp32);
             tcg_temp_free_i32(fp32);
             gen_store_fpr64(ctx, fp64, fd);
@@ -8766,7 +8775,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp32 = tcg_temp_new_i32();
             TCGv_i64 fp64 = tcg_temp_new_i64();
 
-            gen_load_fpr32(fp32, fs);
+            gen_load_fpr32(ctx, fp32, fs);
             gen_helper_float_truncl_s(fp64, cpu_env, fp32);
             tcg_temp_free_i32(fp32);
             gen_store_fpr64(ctx, fp64, fd);
@@ -8780,7 +8789,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp32 = tcg_temp_new_i32();
             TCGv_i64 fp64 = tcg_temp_new_i64();
 
-            gen_load_fpr32(fp32, fs);
+            gen_load_fpr32(ctx, fp32, fs);
             gen_helper_float_ceill_s(fp64, cpu_env, fp32);
             tcg_temp_free_i32(fp32);
             gen_store_fpr64(ctx, fp64, fd);
@@ -8794,7 +8803,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp32 = tcg_temp_new_i32();
             TCGv_i64 fp64 = tcg_temp_new_i64();
 
-            gen_load_fpr32(fp32, fs);
+            gen_load_fpr32(ctx, fp32, fs);
             gen_helper_float_floorl_s(fp64, cpu_env, fp32);
             tcg_temp_free_i32(fp32);
             gen_store_fpr64(ctx, fp64, fd);
@@ -8806,9 +8815,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_helper_float_roundw_s(fp0, cpu_env, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "round.w.s";
@@ -8817,9 +8826,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_helper_float_truncw_s(fp0, cpu_env, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "trunc.w.s";
@@ -8828,9 +8837,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_helper_float_ceilw_s(fp0, cpu_env, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "ceil.w.s";
@@ -8839,9 +8848,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_helper_float_floorw_s(fp0, cpu_env, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "floor.w.s";
@@ -8863,7 +8872,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         break;
     case OPC_MOVCF_S:
         check_insn_opc_removed(ctx, ISA_MIPS32R6);
-        gen_movcf_s(fs, fd, (ft >> 2) & 0x7, ft & 0x1);
+        gen_movcf_s(ctx, fs, fd, (ft >> 2) & 0x7, ft & 0x1);
         opn = "movcf.s";
         break;
     case OPC_MOVZ_S:
@@ -8876,8 +8885,8 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
                 tcg_gen_brcondi_tl(TCG_COND_NE, cpu_gpr[ft], 0, l1);
             }
             fp0 = tcg_temp_new_i32();
-            gen_load_fpr32(fp0, fs);
-            gen_store_fpr32(fp0, fd);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
             gen_set_label(l1);
         }
@@ -8892,8 +8901,8 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             if (ft != 0) {
                 tcg_gen_brcondi_tl(TCG_COND_EQ, cpu_gpr[ft], 0, l1);
                 fp0 = tcg_temp_new_i32();
-                gen_load_fpr32(fp0, fs);
-                gen_store_fpr32(fp0, fd);
+                gen_load_fpr32(ctx, fp0, fs);
+                gen_store_fpr32(ctx, fp0, fd);
                 tcg_temp_free_i32(fp0);
                 gen_set_label(l1);
             }
@@ -8905,9 +8914,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_helper_float_recip_s(fp0, cpu_env, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "recip.s";
@@ -8917,9 +8926,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_helper_float_rsqrt_s(fp0, cpu_env, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "rsqrt.s";
@@ -8930,11 +8939,11 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp0 = tcg_temp_new_i32();
             TCGv_i32 fp1 = tcg_temp_new_i32();
             TCGv_i32 fp2 = tcg_temp_new_i32();
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
-            gen_load_fpr32(fp2, fd);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
+            gen_load_fpr32(ctx, fp2, fd);
             gen_helper_float_maddf_s(fp2, cpu_env, fp0, fp1, fp2);
-            gen_store_fpr32(fp2, fd);
+            gen_store_fpr32(ctx, fp2, fd);
             tcg_temp_free_i32(fp2);
             tcg_temp_free_i32(fp1);
             tcg_temp_free_i32(fp0);
@@ -8947,11 +8956,11 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp0 = tcg_temp_new_i32();
             TCGv_i32 fp1 = tcg_temp_new_i32();
             TCGv_i32 fp2 = tcg_temp_new_i32();
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
-            gen_load_fpr32(fp2, fd);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
+            gen_load_fpr32(ctx, fp2, fd);
             gen_helper_float_msubf_s(fp2, cpu_env, fp0, fp1, fp2);
-            gen_store_fpr32(fp2, fd);
+            gen_store_fpr32(ctx, fp2, fd);
             tcg_temp_free_i32(fp2);
             tcg_temp_free_i32(fp1);
             tcg_temp_free_i32(fp0);
@@ -8962,9 +8971,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         check_insn(ctx, ISA_MIPS32R6);
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_helper_float_rint_s(fp0, cpu_env, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
             opn = "rint.s";
         }
@@ -8973,9 +8982,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         check_insn(ctx, ISA_MIPS32R6);
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_helper_float_class_s(fp0, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
             opn = "class.s";
         }
@@ -8986,10 +8995,10 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp0 = tcg_temp_new_i32();
             TCGv_i32 fp1 = tcg_temp_new_i32();
             TCGv_i32 fp2 = tcg_temp_new_i32();
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
             gen_helper_float_min_s(fp2, cpu_env, fp0, fp1);
-            gen_store_fpr32(fp2, fd);
+            gen_store_fpr32(ctx, fp2, fd);
             tcg_temp_free_i32(fp2);
             tcg_temp_free_i32(fp1);
             tcg_temp_free_i32(fp0);
@@ -9001,11 +9010,11 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
                 TCGv_i32 fp0 = tcg_temp_new_i32();
                 TCGv_i32 fp1 = tcg_temp_new_i32();
 
-                gen_load_fpr32(fp0, fs);
-                gen_load_fpr32(fp1, ft);
+                gen_load_fpr32(ctx, fp0, fs);
+                gen_load_fpr32(ctx, fp1, ft);
                 gen_helper_float_recip2_s(fp0, cpu_env, fp0, fp1);
                 tcg_temp_free_i32(fp1);
-                gen_store_fpr32(fp0, fd);
+                gen_store_fpr32(ctx, fp0, fd);
                 tcg_temp_free_i32(fp0);
             }
             opn = "recip2.s";
@@ -9017,10 +9026,10 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp0 = tcg_temp_new_i32();
             TCGv_i32 fp1 = tcg_temp_new_i32();
             TCGv_i32 fp2 = tcg_temp_new_i32();
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
             gen_helper_float_mina_s(fp2, cpu_env, fp0, fp1);
-            gen_store_fpr32(fp2, fd);
+            gen_store_fpr32(ctx, fp2, fd);
             tcg_temp_free_i32(fp2);
             tcg_temp_free_i32(fp1);
             tcg_temp_free_i32(fp0);
@@ -9031,9 +9040,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             {
                 TCGv_i32 fp0 = tcg_temp_new_i32();
 
-                gen_load_fpr32(fp0, fs);
+                gen_load_fpr32(ctx, fp0, fs);
                 gen_helper_float_recip1_s(fp0, cpu_env, fp0);
-                gen_store_fpr32(fp0, fd);
+                gen_store_fpr32(ctx, fp0, fd);
                 tcg_temp_free_i32(fp0);
             }
             opn = "recip1.s";
@@ -9044,10 +9053,10 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             /* OPC_MAX_S */
             TCGv_i32 fp0 = tcg_temp_new_i32();
             TCGv_i32 fp1 = tcg_temp_new_i32();
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
             gen_helper_float_max_s(fp1, cpu_env, fp0, fp1);
-            gen_store_fpr32(fp1, fd);
+            gen_store_fpr32(ctx, fp1, fd);
             tcg_temp_free_i32(fp1);
             tcg_temp_free_i32(fp0);
             opn = "max.s";
@@ -9057,9 +9066,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             {
                 TCGv_i32 fp0 = tcg_temp_new_i32();
 
-                gen_load_fpr32(fp0, fs);
+                gen_load_fpr32(ctx, fp0, fs);
                 gen_helper_float_rsqrt1_s(fp0, cpu_env, fp0);
-                gen_store_fpr32(fp0, fd);
+                gen_store_fpr32(ctx, fp0, fd);
                 tcg_temp_free_i32(fp0);
             }
             opn = "rsqrt1.s";
@@ -9070,10 +9079,10 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             /* OPC_MAXA_S */
             TCGv_i32 fp0 = tcg_temp_new_i32();
             TCGv_i32 fp1 = tcg_temp_new_i32();
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
             gen_helper_float_maxa_s(fp1, cpu_env, fp0, fp1);
-            gen_store_fpr32(fp1, fd);
+            gen_store_fpr32(ctx, fp1, fd);
             tcg_temp_free_i32(fp1);
             tcg_temp_free_i32(fp0);
             opn = "maxa.s";
@@ -9084,11 +9093,11 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
                 TCGv_i32 fp0 = tcg_temp_new_i32();
                 TCGv_i32 fp1 = tcg_temp_new_i32();
 
-                gen_load_fpr32(fp0, fs);
-                gen_load_fpr32(fp1, ft);
+                gen_load_fpr32(ctx, fp0, fs);
+                gen_load_fpr32(ctx, fp1, ft);
                 gen_helper_float_rsqrt2_s(fp0, cpu_env, fp0, fp1);
                 tcg_temp_free_i32(fp1);
-                gen_store_fpr32(fp0, fd);
+                gen_store_fpr32(ctx, fp0, fd);
                 tcg_temp_free_i32(fp0);
             }
             opn = "rsqrt2.s";
@@ -9100,7 +9109,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp32 = tcg_temp_new_i32();
             TCGv_i64 fp64 = tcg_temp_new_i64();
 
-            gen_load_fpr32(fp32, fs);
+            gen_load_fpr32(ctx, fp32, fs);
             gen_helper_float_cvtd_s(fp64, cpu_env, fp32);
             tcg_temp_free_i32(fp32);
             gen_store_fpr64(ctx, fp64, fd);
@@ -9112,9 +9121,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_helper_float_cvtw_s(fp0, cpu_env, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "cvt.w.s";
@@ -9125,7 +9134,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp32 = tcg_temp_new_i32();
             TCGv_i64 fp64 = tcg_temp_new_i64();
 
-            gen_load_fpr32(fp32, fs);
+            gen_load_fpr32(ctx, fp32, fs);
             gen_helper_float_cvtl_s(fp64, cpu_env, fp32);
             tcg_temp_free_i32(fp32);
             gen_store_fpr64(ctx, fp64, fd);
@@ -9141,8 +9150,8 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp32_0 = tcg_temp_new_i32();
             TCGv_i32 fp32_1 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp32_0, fs);
-            gen_load_fpr32(fp32_1, ft);
+            gen_load_fpr32(ctx, fp32_0, fs);
+            gen_load_fpr32(ctx, fp32_1, ft);
             tcg_gen_concat_i32_i64(fp64, fp32_1, fp32_0);
             tcg_temp_free_i32(fp32_1);
             tcg_temp_free_i32(fp32_0);
@@ -9344,7 +9353,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             gen_load_fpr64(ctx, fp64, fs);
             gen_helper_float_roundw_d(fp32, cpu_env, fp64);
             tcg_temp_free_i64(fp64);
-            gen_store_fpr32(fp32, fd);
+            gen_store_fpr32(ctx, fp32, fd);
             tcg_temp_free_i32(fp32);
         }
         opn = "round.w.d";
@@ -9358,7 +9367,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             gen_load_fpr64(ctx, fp64, fs);
             gen_helper_float_truncw_d(fp32, cpu_env, fp64);
             tcg_temp_free_i64(fp64);
-            gen_store_fpr32(fp32, fd);
+            gen_store_fpr32(ctx, fp32, fd);
             tcg_temp_free_i32(fp32);
         }
         opn = "trunc.w.d";
@@ -9372,7 +9381,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             gen_load_fpr64(ctx, fp64, fs);
             gen_helper_float_ceilw_d(fp32, cpu_env, fp64);
             tcg_temp_free_i64(fp64);
-            gen_store_fpr32(fp32, fd);
+            gen_store_fpr32(ctx, fp32, fd);
             tcg_temp_free_i32(fp32);
         }
         opn = "ceil.w.d";
@@ -9386,7 +9395,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             gen_load_fpr64(ctx, fp64, fs);
             gen_helper_float_floorw_d(fp32, cpu_env, fp64);
             tcg_temp_free_i64(fp64);
-            gen_store_fpr32(fp32, fd);
+            gen_store_fpr32(ctx, fp32, fd);
             tcg_temp_free_i32(fp32);
         }
         opn = "floor.w.d";
@@ -9669,7 +9678,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             gen_load_fpr64(ctx, fp64, fs);
             gen_helper_float_cvts_d(fp32, cpu_env, fp64);
             tcg_temp_free_i64(fp64);
-            gen_store_fpr32(fp32, fd);
+            gen_store_fpr32(ctx, fp32, fd);
             tcg_temp_free_i32(fp32);
         }
         opn = "cvt.s.d";
@@ -9683,7 +9692,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             gen_load_fpr64(ctx, fp64, fs);
             gen_helper_float_cvtw_d(fp32, cpu_env, fp64);
             tcg_temp_free_i64(fp64);
-            gen_store_fpr32(fp32, fd);
+            gen_store_fpr32(ctx, fp32, fd);
             tcg_temp_free_i32(fp32);
         }
         opn = "cvt.w.d";
@@ -9704,9 +9713,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_helper_float_cvts_w(fp0, cpu_env, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "cvt.s.w";
@@ -9717,7 +9726,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp32 = tcg_temp_new_i32();
             TCGv_i64 fp64 = tcg_temp_new_i64();
 
-            gen_load_fpr32(fp32, fs);
+            gen_load_fpr32(ctx, fp32, fs);
             gen_helper_float_cvtd_w(fp64, cpu_env, fp32);
             tcg_temp_free_i32(fp32);
             gen_store_fpr64(ctx, fp64, fd);
@@ -9734,7 +9743,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             gen_load_fpr64(ctx, fp64, fs);
             gen_helper_float_cvts_l(fp32, cpu_env, fp64);
             tcg_temp_free_i64(fp64);
-            gen_store_fpr32(fp32, fd);
+            gen_store_fpr32(ctx, fp32, fd);
             tcg_temp_free_i32(fp32);
         }
         opn = "cvt.s.l";
@@ -9973,7 +9982,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
 
             gen_load_fpr32h(ctx, fp0, fs);
             gen_helper_float_cvts_pu(fp0, cpu_env, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "cvt.s.pu";
@@ -9995,9 +10004,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_helper_float_cvts_pl(fp0, cpu_env, fp0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "cvt.s.pl";
@@ -10008,10 +10017,10 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp0 = tcg_temp_new_i32();
             TCGv_i32 fp1 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
             gen_store_fpr32h(ctx, fp0, fd);
-            gen_store_fpr32(fp1, fd);
+            gen_store_fpr32(ctx, fp1, fd);
             tcg_temp_free_i32(fp0);
             tcg_temp_free_i32(fp1);
         }
@@ -10023,9 +10032,9 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp0 = tcg_temp_new_i32();
             TCGv_i32 fp1 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             gen_load_fpr32h(ctx, fp1, ft);
-            gen_store_fpr32(fp1, fd);
+            gen_store_fpr32(ctx, fp1, fd);
             gen_store_fpr32h(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
             tcg_temp_free_i32(fp1);
@@ -10039,8 +10048,8 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
             TCGv_i32 fp1 = tcg_temp_new_i32();
 
             gen_load_fpr32h(ctx, fp0, fs);
-            gen_load_fpr32(fp1, ft);
-            gen_store_fpr32(fp1, fd);
+            gen_load_fpr32(ctx, fp1, ft);
+            gen_store_fpr32(ctx, fp1, fd);
             gen_store_fpr32h(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
             tcg_temp_free_i32(fp1);
@@ -10055,7 +10064,7 @@  static void gen_farith (DisasContext *ctx, enum fopcode op1,
 
             gen_load_fpr32h(ctx, fp0, fs);
             gen_load_fpr32h(ctx, fp1, ft);
-            gen_store_fpr32(fp1, fd);
+            gen_store_fpr32(ctx, fp1, fd);
             gen_store_fpr32h(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
             tcg_temp_free_i32(fp1);
@@ -10130,7 +10139,7 @@  static void gen_flt3_ldst (DisasContext *ctx, uint32_t opc,
 
             tcg_gen_qemu_ld_tl(t0, t0, ctx->mem_idx, MO_TESL);
             tcg_gen_trunc_tl_i32(fp0, t0);
-            gen_store_fpr32(fp0, fd);
+            gen_store_fpr32(ctx, fp0, fd);
             tcg_temp_free_i32(fp0);
         }
         opn = "lwxc1";
@@ -10162,7 +10171,7 @@  static void gen_flt3_ldst (DisasContext *ctx, uint32_t opc,
         check_cop1x(ctx);
         {
             TCGv_i32 fp0 = tcg_temp_new_i32();
-            gen_load_fpr32(fp0, fs);
+            gen_load_fpr32(ctx, fp0, fs);
             tcg_gen_qemu_st_i32(fp0, t0, ctx->mem_idx, MO_TEUL);
             tcg_temp_free_i32(fp0);
         }
@@ -10219,23 +10228,23 @@  static void gen_flt3_arith (DisasContext *ctx, uint32_t opc,
             tcg_gen_andi_tl(t0, t0, 0x7);
 
             tcg_gen_brcondi_tl(TCG_COND_NE, t0, 0, l1);
-            gen_load_fpr32(fp, fs);
+            gen_load_fpr32(ctx, fp, fs);
             gen_load_fpr32h(ctx, fph, fs);
-            gen_store_fpr32(fp, fd);
+            gen_store_fpr32(ctx, fp, fd);
             gen_store_fpr32h(ctx, fph, fd);
             tcg_gen_br(l2);
             gen_set_label(l1);
             tcg_gen_brcondi_tl(TCG_COND_NE, t0, 4, l2);
             tcg_temp_free(t0);
 #ifdef TARGET_WORDS_BIGENDIAN
-            gen_load_fpr32(fp, fs);
+            gen_load_fpr32(ctx, fp, fs);
             gen_load_fpr32h(ctx, fph, ft);
             gen_store_fpr32h(ctx, fp, fd);
-            gen_store_fpr32(fph, fd);
+            gen_store_fpr32(ctx, fph, fd);
 #else
             gen_load_fpr32h(ctx, fph, fs);
-            gen_load_fpr32(fp, ft);
-            gen_store_fpr32(fph, fd);
+            gen_load_fpr32(ctx, fp, ft);
+            gen_store_fpr32(ctx, fph, fd);
             gen_store_fpr32h(ctx, fp, fd);
 #endif
             gen_set_label(l2);
@@ -10251,13 +10260,13 @@  static void gen_flt3_arith (DisasContext *ctx, uint32_t opc,
             TCGv_i32 fp1 = tcg_temp_new_i32();
             TCGv_i32 fp2 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
-            gen_load_fpr32(fp2, fr);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
+            gen_load_fpr32(ctx, fp2, fr);
             gen_helper_float_madd_s(fp2, cpu_env, fp0, fp1, fp2);
             tcg_temp_free_i32(fp0);
             tcg_temp_free_i32(fp1);
-            gen_store_fpr32(fp2, fd);
+            gen_store_fpr32(ctx, fp2, fd);
             tcg_temp_free_i32(fp2);
         }
         opn = "madd.s";
@@ -10306,13 +10315,13 @@  static void gen_flt3_arith (DisasContext *ctx, uint32_t opc,
             TCGv_i32 fp1 = tcg_temp_new_i32();
             TCGv_i32 fp2 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
-            gen_load_fpr32(fp2, fr);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
+            gen_load_fpr32(ctx, fp2, fr);
             gen_helper_float_msub_s(fp2, cpu_env, fp0, fp1, fp2);
             tcg_temp_free_i32(fp0);
             tcg_temp_free_i32(fp1);
-            gen_store_fpr32(fp2, fd);
+            gen_store_fpr32(ctx, fp2, fd);
             tcg_temp_free_i32(fp2);
         }
         opn = "msub.s";
@@ -10361,13 +10370,13 @@  static void gen_flt3_arith (DisasContext *ctx, uint32_t opc,
             TCGv_i32 fp1 = tcg_temp_new_i32();
             TCGv_i32 fp2 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
-            gen_load_fpr32(fp2, fr);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
+            gen_load_fpr32(ctx, fp2, fr);
             gen_helper_float_nmadd_s(fp2, cpu_env, fp0, fp1, fp2);
             tcg_temp_free_i32(fp0);
             tcg_temp_free_i32(fp1);
-            gen_store_fpr32(fp2, fd);
+            gen_store_fpr32(ctx, fp2, fd);
             tcg_temp_free_i32(fp2);
         }
         opn = "nmadd.s";
@@ -10416,13 +10425,13 @@  static void gen_flt3_arith (DisasContext *ctx, uint32_t opc,
             TCGv_i32 fp1 = tcg_temp_new_i32();
             TCGv_i32 fp2 = tcg_temp_new_i32();
 
-            gen_load_fpr32(fp0, fs);
-            gen_load_fpr32(fp1, ft);
-            gen_load_fpr32(fp2, fr);
+            gen_load_fpr32(ctx, fp0, fs);
+            gen_load_fpr32(ctx, fp1, ft);
+            gen_load_fpr32(ctx, fp2, fr);
             gen_helper_float_nmsub_s(fp2, cpu_env, fp0, fp1, fp2);
             tcg_temp_free_i32(fp0);
             tcg_temp_free_i32(fp1);
-            gen_store_fpr32(fp2, fd);
+            gen_store_fpr32(ctx, fp2, fd);
             tcg_temp_free_i32(fp2);
         }
         opn = "nmsub.s";
@@ -13502,7 +13511,7 @@  static void decode_micromips32_opc (CPUMIPSState *env, DisasContext *ctx,
                 case MOVF_FMT:
                     switch (fmt) {
                     case FMT_SDPS_S:
-                        gen_movcf_s(rs, rt, cc, 0);
+                        gen_movcf_s(ctx, rs, rt, cc, 0);
                         break;
                     case FMT_SDPS_D:
                         gen_movcf_d(ctx, rs, rt, cc, 0);
@@ -13517,7 +13526,7 @@  static void decode_micromips32_opc (CPUMIPSState *env, DisasContext *ctx,
                 case MOVT_FMT:
                     switch (fmt) {
                     case FMT_SDPS_S:
-                        gen_movcf_s(rs, rt, cc, 1);
+                        gen_movcf_s(ctx, rs, rt, cc, 1);
                         break;
                     case FMT_SDPS_D:
                         gen_movcf_d(ctx, rs, rt, cc, 1);
diff --git a/target-mips/translate_init.c b/target-mips/translate_init.c
index 85a65e7..8e088c9 100644
--- a/target-mips/translate_init.c
+++ b/target-mips/translate_init.c
@@ -622,7 +622,8 @@  static const mips_def_t mips_defs[] =
                        (1 << CP0C3_BI) | (1 << CP0C3_ULRI) | (1U << CP0C3_M),
         .CP0_Config4 = MIPS_CONFIG4 | (0xfc << CP0C4_KScrExist) |
                        (3 << CP0C4_IE) | (1 << CP0C4_M),
-        .CP0_Config5_rw_bitmask = (1 << CP0C5_SBRI),
+        .CP0_Config5_rw_bitmask = (1 << CP0C5_SBRI) | (1 << CP0C5_FRE) |
+                                  (1 << CP0C5_UFE),
         .CP0_LLAddr_rw_bitmask = 0,
         .CP0_LLAddr_shift = 0,
         .SYNCI_Step = 32,
@@ -631,9 +632,9 @@  static const mips_def_t mips_defs[] =
         .CP0_PageGrain = (1 << CP0PG_IEC) | (1 << CP0PG_XIE) |
                          (1U << CP0PG_RIE),
         .CP0_PageGrain_rw_bitmask = 0,
-        .CP1_fcr0 = (1 << FCR0_F64) | (1 << FCR0_L) | (1 << FCR0_W) |
-                    (1 << FCR0_D) | (1 << FCR0_S) | (0x00 << FCR0_PRID) |
-                    (0x0 << FCR0_REV),
+        .CP1_fcr0 = (1 << FCR0_FREP) | (1 << FCR0_F64) | (1 << FCR0_L) |
+                    (1 << FCR0_W) | (1 << FCR0_D) | (1 << FCR0_S) |
+                    (0x00 << FCR0_PRID) | (0x0 << FCR0_REV),
         .SEGBITS = 42,
         /* The architectural limit is 59, but we have hardcoded 36 bit
            in some places...