[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH 02/19] fpu/softfloat: Split floatXX_silence_nan
|
From: |
Richard Henderson |
|
Subject: |
Re: [Qemu-devel] [PATCH 02/19] fpu/softfloat: Split floatXX_silence_nan from floatXX_maybe_silence_nan |
|
Date: |
Fri, 11 May 2018 08:16:40 -0700 |
|
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.7.0 |
On 05/11/2018 08:06 AM, Alex Bennée wrote:
> but...but...
>>> -float16 float16_maybe_silence_nan(float16 a_, float_status *status)
>>> +
>>> +float16 float16_maybe_silence_nan(float16 a, float_status *status)
>>> {
>>> + if (float16_is_signaling_nan(a, status)) {
>>> + float16_silence_nan(a, status);
>
> You need a return here.
Oops.
> Amazing what tests pass with those in, but RISU picked them up.
Heh. In my case I hadn't bothered testing until further down the patch set
when these functions have been removed.
r~
- [Qemu-devel] [PATCH 00/19] softfloat: Clean up NaN handling, Richard Henderson, 2018/05/10
- [Qemu-devel] [PATCH 04/19] fpu/softfloat: Canonicalize NaN fraction, Richard Henderson, 2018/05/10
- [Qemu-devel] [PATCH 03/19] fpu/softfloat: Move softfloat-specialize.h below FloatParts definition, Richard Henderson, 2018/05/10
- [Qemu-devel] [PATCH 05/19] fpu/softfloat: Introduce parts_is_snan_frac, Richard Henderson, 2018/05/10
- [Qemu-devel] [PATCH 06/19] fpu/softfloat: Replace float_class_dnan with parts_default_nan, Richard Henderson, 2018/05/10
- [Qemu-devel] [PATCH 08/19] target/arm: Use floatX_silence_nan when we have already checked for SNaN, Richard Henderson, 2018/05/10
- [Qemu-devel] [PATCH 09/19] target/arm: Remove floatX_maybe_silence_nan from conversions, Richard Henderson, 2018/05/10
- [Qemu-devel] [PATCH 07/19] fpu/softfloat: Replace float_class_msnan with parts_silence_nan, Richard Henderson, 2018/05/10