With Debian package beignet-opencl-icd 1.2.1-1, Tesseract + OpenCL fails:
[DS] Profile file not available (tesseract_opencl_profile_devices.dat); performing profiling.
[DS] Device: "Intel(R) HD Graphics IvyBridge M GT2" (OpenCL) evaluation...
ASSERTION FAILED: sel.hasDoubleType()
at file /home/geier/beignet/backend/src/backend/gen_insn_selection.cpp, function void gbe::ConvertInstructionPattern::convertDoubleToSmallInts(gbe::Selection::Opaque&, const gbe::ir::ConvertInstruction&, bool&) const, line 5269
Trace/breakpoint trap
Using a pure float expression (instead of double) fixes this issue.
Signed-off-by: Stefan Weil <sw@weilnetz.de>
Cmake builds need the additional directories `arch` and `lstm`
to find all include files and also for linking.
Signed-off-by: Stefan Weil <sw@weilnetz.de>
gcc report:
ccstruct/blamer.cpp:343:65: warning:
'truth_x' may be used uninitialized in this function [-Wmaybe-uninitialized]
Signed-off-by: Stefan Weil <sw@weilnetz.de>
gcc report:
lstm/lstmrecognizer.cpp:608:47: error: 'isnan' was not declared in this scope
ASSERT_HOST(!isnan(output.f(t)[null_char_]));
Signed-off-by: Stefan Weil <sw@weilnetz.de>
training/commontraining.cpp:824:3: warning:
'register' storage class specifier is deprecated and incompatible with C++1z [-Wdeprecated-register]
...
Signed-off-by: Stefan Weil <sw@weilnetz.de>
CL_PROGRAM_NUM_DEVICES expects a cl_uint.
Passing size_t results in a wrong value for numDevices on hosts where
sizeof(cl_uint) != sizeof(size_t).
Signed-off-by: Stefan Weil <sw@weilnetz.de>
CL_CONTEXT_NUM_DEVICES expects a cl_uint.
Passing size_t results in a wrong value for numDevices on hosts where
sizeof(cl_uint) != sizeof(size_t). This results in errors like these:
Tesseract Open Source OCR Engine v3.05.00dev with Leptonica
OpenCL error code is -44 at when clCreateKernel kernel_HistogramRectAllChannels .
OpenCL error code is -44 at when clCreateKernel kernel_HistogramRectAllChannelsReduction .
OpenCL error code is -48 at when clSetKernelArg imageBuffer .
...
Signed-off-by: Stefan Weil <sw@weilnetz.de>
The compare method is called very often, so even small improvements
are important.
The new code avoids one comparison in each loop iteration.
This results in smaller code (60 bytes for x86_64, gcc).
Signed-off-by: Stefan Weil <sw@weilnetz.de>
