diff options
Diffstat (limited to 'src/codegen_arm.h')
| -rw-r--r-- | src/codegen_arm.h | 225 |
1 files changed, 225 insertions, 0 deletions
diff --git a/src/codegen_arm.h b/src/codegen_arm.h new file mode 100644 index 0000000..3493a11 --- /dev/null +++ b/src/codegen_arm.h @@ -0,0 +1,225 @@ +/* + + This file is part of FFTS -- The Fastest Fourier Transform in the South + + Copyright (c) 2012, Anthony M. Blake <amb@anthonix.com> + Copyright (c) 2012, The University of Waikato + + All rights reserved. + + Redistribution and use in source and binary forms, with or without + modification, are permitted provided that the following conditions are met: + * Redistributions of source code must retain the above copyright + notice, this list of conditions and the following disclaimer. + * Redistributions in binary form must reproduce the above copyright + notice, this list of conditions and the following disclaimer in the + documentation and/or other materials provided with the distribution. + * Neither the name of the organization nor the + names of its contributors may be used to endorse or promote products + derived from this software without specific prior written permission. + + THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND + ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED + WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE + DISCLAIMED. IN NO EVENT SHALL ANTHONY M. BLAKE BE LIABLE FOR ANY + DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES + (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; + LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT + (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS + SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. + +*/ + +#ifndef __CODEGEN_ARM_H__ +#define __CODEGEN_ARM_H__ + + + +uint32_t BL(void *pos, void *target) { + return 0xeb000000 | (((target - pos) / 4) & 0xffffff); +} + +uint32_t B(uint8_t r) { + return 0xe12fff10 | r; +} + +uint32_t MOV(uint8_t dst, uint8_t src) { + return 0xe1a00000 | (src & 0xf) | ((dst & 0xf) << 12); +} + +void ADDI(uint32_t **p, uint8_t dst, uint8_t src, int32_t imm) { + int32_t oimm = imm; + if(imm < 0) { + imm = -imm; + uint32_t shamt = (__builtin_ctzl(imm)>23)?23:__builtin_ctzl(imm); + if(shamt & 1) shamt -= 1; + imm >>= shamt; + shamt = (32 - shamt)/2; + + // if(imm > 255) fprintf(stderr, "imm>255: %d\n", oimm); + *(*p)++ = 0xe2400000 | ((src & 0xf) << 16) | ((dst & 0xf) << 12) | ((shamt & 0xf) << 8) | (imm & 0xff); + + if(imm > 255) ADDI(p, dst, src, (oimm + ((imm & 0xff) << (32-shamt*2)))); + + }else{ + uint32_t shamt = (__builtin_ctzl(imm)>23)?23:__builtin_ctzl(imm); + if(shamt & 1) shamt -= 1; + imm >>= shamt; + shamt = (32 - shamt)/2; + +// if(imm > 255) fprintf(stderr, "imm>255: %d\n", oimm); + + *(*p)++ = 0xe2800000 | ((src & 0xf) << 16) | ((dst & 0xf) << 12) | ((shamt & 0xf) << 8) | (imm & 0xff); + + if(imm > 255) ADDI(p, dst, src, (oimm - ((imm & 0xff) << (32-shamt*2)))); + } +} + +uint32_t LDRI(uint8_t dst, uint8_t base, uint32_t offset) { + return 0xe5900000 | ((dst & 0xf) << 12) + | ((base & 0xf) << 16) | (offset & 0xfff) ; +} + +void MOVI(uint32_t **p, uint8_t dst, uint32_t imm) { + uint32_t oimm = imm; + + uint32_t shamt = (__builtin_ctzl(imm)>23)?23:__builtin_ctzl(imm); + if(shamt & 1) shamt -= 1; + imm >>= shamt; + shamt = (32 - shamt)/2; + *(*p)++ = 0xe3a00000 | ((dst & 0xf) << 12) | ((shamt & 0xf) << 8) | (imm & 0xff) ; + if(imm > 255) ADDI(p, dst, dst, (oimm - ((imm & 0xff) << (32-shamt*2)))); +} + +uint32_t PUSH_LR() { return 0xe92d4ff0; } //0xe92d4000; } +uint32_t POP_LR() { return 0xe8bd8ff0; } //0xe8bd8000; } + +static FFTS_INLINE insns_t* generate_size4_base_case(insns_t **fp, int sign) +{ + insns_t *x_4_addr; + size_t len; + + x_4_addr = *fp; + +#ifdef HAVE_NEON + len = (char*) neon_x8 - (char*) neon_x4; + memcpy(x_4_addr, neon_x4, len); + + if (sign < 0) { + x_4_addr[26] ^= 0x00200000; + x_4_addr[28] ^= 0x00200000; + x_4_addr[31] ^= 0x00200000; + x_4_addr[32] ^= 0x00200000; + } +#else + len = (char*) vfp_x8 - (char*) vfp_x4; + memcpy(x_4_addr, vfp_x4, len); + + if (sign > 0) { + x_4_addr[36] ^= 0x00000040; + x_4_addr[38] ^= 0x00000040; + x_4_addr[43] ^= 0x00000040; + x_4_addr[44] ^= 0x00000040; + } +#endif + + *fp += len / 4; + return x_4_addr; +} + +static FFTS_INLINE insns_t* generate_size8_base_case(insns_t **fp, int sign) +{ + insns_t *x_8_addr; + ptrdiff_t len; + + x_8_addr = *fp; + +#ifdef HAVE_NEON + len = (char*) neon_x8_t - (char*) neon_x8; + memcpy(x_8_addr, neon_x8, len); + + /* + * Changes adds to subtracts and vice versa to allow the computation + * of both the IFFT and FFT + */ + if (sign < 0) { + x_8_addr[31] ^= 0x00200000; + x_8_addr[32] ^= 0x00200000; + x_8_addr[33] ^= 0x00200000; + x_8_addr[34] ^= 0x00200000; + x_8_addr[65] ^= 0x00200000; + x_8_addr[66] ^= 0x00200000; + x_8_addr[70] ^= 0x00200000; + x_8_addr[74] ^= 0x00200000; + x_8_addr[97] ^= 0x00200000; + x_8_addr[98] ^= 0x00200000; + x_8_addr[102] ^= 0x00200000; + x_8_addr[104] ^= 0x00200000; + } + + *fp += len / 4; + + //uint32_t *x_8_t_addr = fp; + //memcpy(fp, neon_x8_t, neon_end - neon_x8_t); + //fp += (neon_end - neon_x8_t) / 4; +#else + len = (char*) vfp_end - (char*) vfp_x8; + memcpy(x_8_addr, vfp_x8, len); + + if (sign > 0) { + x_8_addr[65] ^= 0x00000040; + x_8_addr[66] ^= 0x00000040; + x_8_addr[68] ^= 0x00000040; + x_8_addr[70] ^= 0x00000040; + x_8_addr[103] ^= 0x00000040; + x_8_addr[104] ^= 0x00000040; + x_8_addr[105] ^= 0x00000040; + x_8_addr[108] ^= 0x00000040; + x_8_addr[113] ^= 0x00000040; + x_8_addr[114] ^= 0x00000040; + x_8_addr[117] ^= 0x00000040; + x_8_addr[118] ^= 0x00000040; + } + + *fp += len / 4; +#endif + + return x_8_addr; +} + +static FFTS_INLINE insns_t* generate_prologue(insns_t **fp, ffts_plan_t *p) +{ + insns_t *start = fp; + + *(*fp)++ = PUSH_LR(); + *(*fp)++ = 0xed2d8b10; + + ADDI(fp, 3, 1, 0); + ADDI(fp, 7, 1, N); + ADDI(fp, 5, 1, 2*N); + ADDI(fp, 10, 7, 2*N); + ADDI(fp, 4, 5, 2*N); + ADDI(fp, 8, 10, 2*N); + ADDI(fp, 6, 4, 2*N); + ADDI(fp, 9, 8, 2*N); + + // load offsets into r12 + *(*fp)++ = LDRI(12, 0, ((uint32_t) &p->offsets) - ((uint32_t) p)); + // *(*fp)++ = LDRI(1, 0, 4); // load ws into r1 + ADDI(fp, 1, 0, 0); + + ADDI(fp, 0, 2, 0), // mov out into r0 + *(*fp)++ = LDRI(2, 1, ((uint32_t) &p->ee_ws) - ((uint32_t) p)); + +#ifdef HAVE_NEON + MOVI(fp, 11, p->i0); +#else + MOVI(fp, 11, p->i0); +#endif + + return start; +} + +#endif +// vim: set autoindent noexpandtab tabstop=3 shiftwidth=3: |
