[Beignet] [PATCH] utest_pow: don't fail on declared lack of denormals

[Zhigang Gong]

Pushed, thanks for the patch and review.

On Thu, Apr 23, 2015 at 04:26:30AM +0000, Song, Ruiling wrote:
> LGTM
> 
> > -----Original Message-----
> > From: Beignet [mailto:beignet-bounces at lists.freedesktop.org] On Behalf Of
> > Rebecca N. Palmer
> > Sent: Saturday, April 18, 2015 5:19 AM
> > To: beignet at lists.freedesktop.org
> > Subject: [Beignet] [PATCH] utest_pow: don't fail on declared lack of
> > denormals
> > 
> > (Resend: the first copy appears to have gone missing)
> > 
> > 0.01**20.5 is denormal; at least Ivy Bridge does not support denormals and
> > hence returns 0.  As this is allowed by the OpenCL standard, it shouldn't fail
> > the test.
> > 
> > Signed-off-by: Rebecca Palmer <rebecca_palmer at zoho.com>
> > 
> > diff --git a/utests/builtin_pow.cpp b/utests/builtin_pow.cpp index
> > a18f31e..6e0f8dd 100644
> > --- a/utests/builtin_pow.cpp
> > +++ b/utests/builtin_pow.cpp
> > @@ -37,6 +37,9 @@ static void builtin_pow(void)
> >         input_data2[i*count_input_ori+k] = ori_data[k];
> >       }
> > 
> > +  cl_device_fp_config fp_config;
> > +  clGetDeviceInfo(device, CL_DEVICE_SINGLE_FP_CONFIG,
> > + sizeof(cl_device_fp_config), &fp_config, 0);  bool denormals_supported
> > + = fp_config & CL_FP_DENORM;
> >     const char* env_strict = getenv("OCL_STRICT_CONFORMANCE");
> >     float ULPSIZE_FACTOR = 16.0;
> >     if (env_strict == NULL || strcmp(env_strict, "0") == 0) @@ -75,7 +78,8
> > @@ static void builtin_pow(void)
> >   #if udebug
> >         if ( (isinf(cpu_data[index_cur]) && !isinf(gpu_data[index_cur])) ||
> >              (isnan(cpu_data[index_cur]) && !isnan(gpu_data[index_cur]))
> > ||
> > -           (fabs(gpu_data[index_cur] - cpu_data[index_cur]) >
> > cl_FLT_ULP(cpu_data[index_cur]) * ULPSIZE_FACTOR)   )
> > +           (fabs(gpu_data[index_cur] - cpu_data[index_cur]) >
> > cl_FLT_ULP(cpu_data[index_cur]) * ULPSIZE_FACTOR
> > +           && (denormals_supported || gpu_data[index_cur]!=0 ||
> > + std::fpclassify(cpu_data[index_cur])!=FP_SUBNORMAL) ) )
> >         {
> >           printf_c("%d/%d: x:%f, y:%f -> gpu:%f  cpu:%f\n", k, i,
> > input_data1[k], input_data2[k], gpu_data[index_cur], cpu_data[index_cur]);
> >         }
> > @@ -88,7 +92,8 @@ static void builtin_pow(void)
> >          OCL_ASSERT(isnan(gpu_data[index_cur]));
> >        else
> >        {
> > -       OCL_ASSERT(fabs(gpu_data[index_cur] - cpu_data[index_cur]) <
> > cl_FLT_ULP(cpu_data[index_cur]) * ULPSIZE_FACTOR);
> > +       OCL_ASSERT((fabs(gpu_data[index_cur] - cpu_data[index_cur]) <
> > cl_FLT_ULP(cpu_data[index_cur]) * ULPSIZE_FACTOR) ||
> > +       (!denormals_supported && gpu_data[index_cur]==0 &&
> > + std::fpclassify(cpu_data[index_cur])==FP_SUBNORMAL) );
> >        }
> >   #endif
> >       }
> > 
> > 
> > 
> > _______________________________________________
> > Beignet mailing list
> > Beignet at lists.freedesktop.org
> > http://lists.freedesktop.org/mailman/listinfo/beignet
> _______________________________________________
> Beignet mailing list
> Beignet at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/beignet