mirror of
https://github.com/jart/cosmopolitan.git
synced 2025-01-31 19:43:32 +00:00
f4f4caab0e
I wanted a tiny scriptable meltdown proof way to run userspace programs and visualize how program execution impacts memory. It helps to explain how things like Actually Portable Executable works. It can show you how the GCC generated code is going about manipulating matrices and more. I didn't feel fully comfortable with Qemu and Bochs because I'm not smart enough to understand them. I wanted something like gVisor but with much stronger levels of assurances. I wanted a single binary that'll run, on all major operating systems with an embedded GPL barrier ZIP filesystem that is tiny enough to transpile to JavaScript and run in browsers too. https://justine.storage.googleapis.com/emulator625.mp4
67 lines
4.2 KiB
C
67 lines
4.2 KiB
C
#ifndef COSMOPOLITAN_TOOL_VIZ_LIB_CONVOLUTION_H_
|
|
#define COSMOPOLITAN_TOOL_VIZ_LIB_CONVOLUTION_H_
|
|
#include "dsp/core/q.h"
|
|
#if !(__ASSEMBLER__ + __LINKER__ + 0)
|
|
/* clang-format off */
|
|
|
|
#define CONVROUND(X) ((int)rint(X))
|
|
|
|
#define CONVOLVE1X1(COEFFICIENT, DY, DX) (COEFFICIENT * (img[iy[y DY]][ix[x DX]]))
|
|
#define CONVOLVE1X12(IMG, COEFFICIENT, DY, DX) (COEFFICIENT * IMG[iy[y DY]][ix[x DX]])
|
|
|
|
#define CONVOLVE3X3(NORMALIZER, \
|
|
C00, C01, C02, \
|
|
C10, C11, C12, \
|
|
C20, C21, C22) \
|
|
(NORMALIZER * \
|
|
(CONVOLVE1X1(C00,-1,-1)+CONVOLVE1X1(C01,-1,+0)+CONVOLVE1X1(C02,-1,+1)+ \
|
|
CONVOLVE1X1(C10,+0,-1)+CONVOLVE1X1(C11,+0,+0)+CONVOLVE1X1(C12,+0,+1)+ \
|
|
CONVOLVE1X1(C20,+1,-1)+CONVOLVE1X1(C21,+1,+0)+CONVOLVE1X1(C22,+1,+1)))
|
|
|
|
#define CONVOLVE3X6(NORMALIZER, \
|
|
C00, C01, C02, \
|
|
C10, C11, C12, \
|
|
C20, C21, C22) \
|
|
(NORMALIZER * \
|
|
(CONVOLVE1X1(C00/2.f,-1,-1)+CONVOLVE1X1(C01/2.f,-1,+0)+CONVOLVE1X1(C02/2.f,-1,+1)+ \
|
|
CONVOLVE1X1(C00/2.f,-1,-1)+CONVOLVE1X1(C01/2.f,-1,+0)+CONVOLVE1X1(C02/2.f,-1,+1)+ \
|
|
CONVOLVE1X1(C10/2.f,+0,-1)+CONVOLVE1X1(C11/2.f,+0,+0)+CONVOLVE1X1(C12/2.f,+0,+1)+ \
|
|
CONVOLVE1X1(C10/2.f,+0,-1)+CONVOLVE1X1(C11/2.f,+0,+0)+CONVOLVE1X1(C12/2.f,+0,+1)+ \
|
|
CONVOLVE1X1(C20/2.f,+1,-1)+CONVOLVE1X1(C21/2.f,+1,+0)+CONVOLVE1X1(C22/2.f,+1,+1)+ \
|
|
CONVOLVE1X1(C20/2.f,+1,-1)+CONVOLVE1X1(C21/2.f,+1,+0)+CONVOLVE1X1(C22/2.f,+1,+1)))
|
|
|
|
/* clang-format on */
|
|
|
|
#define CONVOLVE5X5(M, NORM, IMG, C00, C01, C02, C03, C04, C10, C11, C12, C13, \
|
|
C14, C20, C21, C22, C23, C24, C30, C31, C32, C33, C34, \
|
|
C40, C41, C42, C43, C44) \
|
|
QRS(M, (CONVOLVE1X12(IMG, CONVROUND((NORM * C00) * (1 << M)), -2, -2) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C01) * (1 << M)), -2, -1) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C02) * (1 << M)), -2, +0) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C03) * (1 << M)), -2, +1) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C04) * (1 << M)), -2, +2) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C10) * (1 << M)), -1, -2) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C11) * (1 << M)), -1, -1) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C12) * (1 << M)), -1, +0) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C13) * (1 << M)), -1, +1) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C14) * (1 << M)), -1, +2) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C20) * (1 << M)), +0, -2) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C21) * (1 << M)), +0, -1) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C22) * (1 << M)), +0, +0) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C23) * (1 << M)), +0, +1) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C24) * (1 << M)), +0, +2) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C30) * (1 << M)), +1, -2) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C31) * (1 << M)), +1, -1) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C32) * (1 << M)), +1, +0) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C33) * (1 << M)), +1, +1) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C34) * (1 << M)), +1, +2) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C40) * (1 << M)), +2, -2) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C41) * (1 << M)), +2, -1) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C42) * (1 << M)), +2, +0) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C43) * (1 << M)), +2, +1) + \
|
|
CONVOLVE1X12(IMG, CONVROUND((NORM * C44) * (1 << M)), +2, +2)))
|
|
|
|
unsigned *convoindex(unsigned, unsigned, unsigned) mallocesque;
|
|
|
|
#endif /* !(__ASSEMBLER__ + __LINKER__ + 0) */
|
|
#endif /* COSMOPOLITAN_TOOL_VIZ_LIB_CONVOLUTION_H_ */
|