[PATCH] ARM: VFP: Fix emulation of multiply accumulate instructions

[PATCH] ARM: VFP: Fix emulation of multiply accumulate instructions

[Jay Foad]

The emulation for single and double precision multiply accumulate
instructions correctly normalised any denormal values in the operand
registers, but failed to normalise the destination (accumulator)
register.

This fixes https://bugzilla.kernel.org/show_bug.cgi?id=70501

Signed-off-by: Jay Foad <jay.foad@gmail.com>
---
 arch/arm/vfp/vfpdouble.c | 2 ++
 arch/arm/vfp/vfpsingle.c | 2 ++
 2 files changed, 4 insertions(+)

diff --git a/arch/arm/vfp/vfpdouble.c b/arch/arm/vfp/vfpdouble.c
index 6cac43bd1d86..423f56dd4028 100644
--- a/arch/arm/vfp/vfpdouble.c
+++ b/arch/arm/vfp/vfpdouble.c
@@ -866,6 +866,8 @@ vfp_double_multiply_accumulate(int dd, int dn, int dm, u32 fpscr, u32 negate, ch
 		vdp.sign = vfp_sign_negate(vdp.sign);
 
 	vfp_double_unpack(&vdn, vfp_get_double(dd));
+	if (vdn.exponent == 0 && vdn.significand)
+		vfp_double_normalise_denormal(&vdn);
 	if (negate & NEG_SUBTRACT)
 		vdn.sign = vfp_sign_negate(vdn.sign);
 
diff --git a/arch/arm/vfp/vfpsingle.c b/arch/arm/vfp/vfpsingle.c
index b252631b406b..4f96c1617aae 100644
--- a/arch/arm/vfp/vfpsingle.c
+++ b/arch/arm/vfp/vfpsingle.c
@@ -915,6 +915,8 @@ vfp_single_multiply_accumulate(int sd, int sn, s32 m, u32 fpscr, u32 negate, cha
 	v = vfp_get_float(sd);
 	pr_debug("VFP: s%u = %08x\n", sd, v);
 	vfp_single_unpack(&vsn, v);
+	if (vsn.exponent == 0 && vsn.significand)
+		vfp_single_normalise_denormal(&vsn);
 	if (negate & NEG_SUBTRACT)
 		vsn.sign = vfp_sign_negate(vsn.sign);
 
-- 
1.8.3.2

[PATCH] ARM: VFP: Fix emulation of multiply accumulate instructions

[Dave Martin]

On Wed, Mar 26, 2014 at 04:49:09PM +0000, Jay Foad wrote:
> The emulation for single and double precision multiply accumulate
> instructions correctly normalised any denormal values in the operand
> registers, but failed to normalise the destination (accumulator)
> register.
> 
> This fixes https://bugzilla.kernel.org/show_bug.cgi?id=70501
> 
> Signed-off-by: Jay Foad <jay.foad@gmail.com>

Reviewed-by: Dave Martin <Dave.Martin@arm.com>

This appears to be clean and straightforward, so if nobody else has
commented on this by the time -rc1 comes out, I suggest you rebase it
and send it to the patch system.  (See
http://www.arm.linux.org.uk/developer/patches/info.php for details
-- make sure you read it carefully.)

Has this issue been hitting real-world users of the kernel, or is
this fixing a latent bug?

Cheers
---Dave

> ---
>  arch/arm/vfp/vfpdouble.c | 2 ++
>  arch/arm/vfp/vfpsingle.c | 2 ++
>  2 files changed, 4 insertions(+)
> 
> diff --git a/arch/arm/vfp/vfpdouble.c b/arch/arm/vfp/vfpdouble.c
> index 6cac43bd1d86..423f56dd4028 100644
> --- a/arch/arm/vfp/vfpdouble.c
> +++ b/arch/arm/vfp/vfpdouble.c
> @@ -866,6 +866,8 @@ vfp_double_multiply_accumulate(int dd, int dn, int dm, u32 fpscr, u32 negate, ch
>  		vdp.sign = vfp_sign_negate(vdp.sign);
>  
>  	vfp_double_unpack(&vdn, vfp_get_double(dd));
> +	if (vdn.exponent == 0 && vdn.significand)
> +		vfp_double_normalise_denormal(&vdn);
>  	if (negate & NEG_SUBTRACT)
>  		vdn.sign = vfp_sign_negate(vdn.sign);
>  
> diff --git a/arch/arm/vfp/vfpsingle.c b/arch/arm/vfp/vfpsingle.c
> index b252631b406b..4f96c1617aae 100644
> --- a/arch/arm/vfp/vfpsingle.c
> +++ b/arch/arm/vfp/vfpsingle.c
> @@ -915,6 +915,8 @@ vfp_single_multiply_accumulate(int sd, int sn, s32 m, u32 fpscr, u32 negate, cha
>  	v = vfp_get_float(sd);
>  	pr_debug("VFP: s%u = %08x\n", sd, v);
>  	vfp_single_unpack(&vsn, v);
> +	if (vsn.exponent == 0 && vsn.significand)
> +		vfp_single_normalise_denormal(&vsn);
>  	if (negate & NEG_SUBTRACT)
>  		vsn.sign = vfp_sign_negate(vsn.sign);
>  
> -- 
> 1.8.3.2
> 
> 
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

[PATCH] ARM: VFP: Fix emulation of multiply accumulate instructions

[Jay Foad]

On 8 April 2014 14:13, Dave Martin <Dave.Martin@arm.com> wrote:
> On Wed, Mar 26, 2014 at 04:49:09PM +0000, Jay Foad wrote:
>> The emulation for single and double precision multiply accumulate
>> instructions correctly normalised any denormal values in the operand
>> registers, but failed to normalise the destination (accumulator)
>> register.
>>
>> This fixes https://bugzilla.kernel.org/show_bug.cgi?id=70501
>>
>> Signed-off-by: Jay Foad <jay.foad@gmail.com>
>
> Reviewed-by: Dave Martin <Dave.Martin@arm.com>
>
> This appears to be clean and straightforward, so if nobody else has
> commented on this by the time -rc1 comes out, I suggest you rebase it
> and send it to the patch system.  (See
> http://www.arm.linux.org.uk/developer/patches/info.php for details
> -- make sure you read it carefully.)

Done.

> Has this issue been hitting real-world users of the kernel, or is
> this fixing a latent bug?

Real world: it was causing failures in the test suite for an APL
interpreter running on the Raspberry Pi.

Thanks,
Jay.