mirrors/opencv
2
0
Fork 0
mirror of https://github.com/opencv/opencv.git synced 2026-09-25 04:09:57 +03:00
Code Issues Packages Projects Releases Wiki Activity

activation is nullptr

Browse Source
This commit is contained in:
velonica0
2026-08-04 20:27:23 -07:00
parent 484dc43c22
commit 515ad7725a
1 changed files with 22 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files
modules/dnn/src/layers/cpu_kernels/conv2_depthwise.simd.hpp
+22 -5
View File
@@ -40,10 +40,12 @@ static void depthwiseConv32f(const void* inp__, const void* residual__,
parallel_for_(Range(0, NC1), [&](const Range& range)
{
// Offer this task range to an accelerated HAL first, flattening the descriptor into a
// stable C argument list (no dnn types cross the boundary). The generic activation is a
// function pointer that cannot cross the ABI, so only the fast-activation path is offered
// (out = min(s>=0 ? s : s*alpha, maxval)); on NOT_IMPLEMENTED fall through to the built-in.
if (cs.activation == nullptr)
// stable C argument list (no dnn types cross the boundary). The HAL fuses conv, scale,
// bias, residual and the fast-activation (out = min(s>=0 ? s : s*alpha, maxval)). A
// generic activation is a function pointer that cannot cross the ABI, so when one is
// present we still run the HAL for the heavy conv and apply the (elementwise) activation
// over this range afterwards -- matching the built-in fast-epilogue-then-activation
// order below. On NOT_IMPLEMENTED fall through to the built-in path.
{
int sd = cs.nspatialdims;
int insize[3] = { sd > 2 ? cs.inpshape[sd-1] : 1, sd > 1 ? cs.inpshape[sd] : 1, cs.inpshape[sd+1] };
@@ -57,7 +59,7 @@ static void depthwiseConv32f(const void* inp__, const void* residual__,
case FAST_ACTIV_NONE: default_alpha = 1.f; break;
default: break; // FAST_ACTIV_RELU: maxval = FLT_MAX, default_alpha = 0
}
CALL_HAL(dnn_depthwise_conv32f, cv_hal_dnn_depthwise_conv32f,
int hal_res = cv_hal_dnn_depthwise_conv32f(
(const float*)inp__, (const float*)residual__, (float*)out__,
(const float*)weights__, scale__, bias__,
cs.inpshape.C, cs.inpshape.back(), cs.inpshape[1],
@@ -65,6 +67,21 @@ static void depthwiseConv32f(const void* inp__, const void* residual__,
cs.coordtab.data(), cs.ofstab.data(), (int)cs.ofstab.size(),
maxval, default_alpha, prelu_slope,
range.start, range.end);
if (hal_res == CV_HAL_ERROR_OK) {
if (cs.activation) {
// The HAL wrote conv + scale/bias/residual with an identity fast-epilogue;
// apply the generic activation in place over this range's output planes.
int planesz = outsize[0]*outsize[1]*outsize[2]*cs.inpshape.back();
const float* activParams = cs.activParams.data();
float* outp = (float*)out__ + range.start*planesz;
for (int nc1 = range.start; nc1 < range.end; nc1++, outp += planesz)
cs.activation(outp, outp, planesz, activParams);
}
return;
}
else if (hal_res != CV_HAL_ERROR_NOT_IMPLEMENTED)
CV_Error_(cv::Error::StsInternal,
("HAL implementation dnn_depthwise_conv32f ==> cv_hal_dnn_depthwise_conv32f returned %d (0x%08x)", hal_res, hal_res));
}
constexpr int MAX_CONV_DIMS = ConvState::MAX_CONV_DIMS;