From: Georgi Gerganov Date: Mon, 10 Apr 2023 19:39:07 +0000 (+0300) Subject: ggml : optimize ggml_cpy() for contiguous dst X-Git-Tag: upstream/0.0.1642~1559 X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=8f291d9fc91e53cd114c11153b4537c08f8ec863;p=pkg%2Fggml%2Fsources%2Fggml ggml : optimize ggml_cpy() for contiguous dst --- diff --git a/src/ggml.c b/src/ggml.c index a8f36e64..3942379e 100644 --- a/src/ggml.c +++ b/src/ggml.c @@ -4885,6 +4885,85 @@ static void ggml_compute_forward_dup_f16( // TODO: add more special-case implementations for tensor shapes/strides that can benefit from memcpy + if (ggml_is_contiguous(dst)) { + if (src0->nb[0] == sizeof(ggml_fp16_t)) { + if (dst->type == GGML_TYPE_F16) { + size_t id = 0; + const size_t rs = ne00*nb00; + + for (int i03 = 0; i03 < ne03; i03++) { + for (int i02 = 0; i02 < ne02; i02++) { + for (int i01 = 0; i01 < ne01; i01++) { + const char * src0_ptr = (char *) src0->data + i01*nb01 + i02*nb02 + i03*nb03; + char * dst_ptr = (char *) dst->data + id*rs; + + memcpy(dst_ptr, src0_ptr, rs); + + id++; + } + } + } + } else if (dst->type == GGML_TYPE_F32) { + size_t id = 0; + float * dst_ptr = (float *) dst->data; + + for (int i03 = 0; i03 < ne03; i03++) { + for (int i02 = 0; i02 < ne02; i02++) { + for (int i01 = 0; i01 < ne01; i01++) { + for (int i00 = 0; i00 < ne00; i00++) { + const ggml_fp16_t * src0_ptr = (ggml_fp16_t *) ((char *) src0->data + i00*nb00 + i01*nb01 + i02*nb02 + i03*nb03); + + dst_ptr[id] = GGML_FP16_TO_FP32(*src0_ptr); + id++; + } + } + } + } + } else { + GGML_ASSERT(false); // TODO: implement + } + } else { + //printf("%s: this is not optimal - fix me\n", __func__); + + if (dst->type == GGML_TYPE_F32) { + size_t id = 0; + float * dst_ptr = (float *) dst->data; + + for (int i03 = 0; i03 < ne03; i03++) { + for (int i02 = 0; i02 < ne02; i02++) { + for (int i01 = 0; i01 < ne01; i01++) { + for (int i00 = 0; i00 < ne00; i00++) { + const ggml_fp16_t * src0_ptr = (ggml_fp16_t *) ((char *) src0->data + i00*nb00 + i01*nb01 + i02*nb02 + i03*nb03); + + dst_ptr[id] = GGML_FP16_TO_FP32(*src0_ptr); + id++; + } + } + } + } + } else if (dst->type == GGML_TYPE_F16) { + size_t id = 0; + ggml_fp16_t * dst_ptr = (ggml_fp16_t *) dst->data; + + for (int i03 = 0; i03 < ne03; i03++) { + for (int i02 = 0; i02 < ne02; i02++) { + for (int i01 = 0; i01 < ne01; i01++) { + for (int i00 = 0; i00 < ne00; i00++) { + const ggml_fp16_t * src0_ptr = (ggml_fp16_t *) ((char *) src0->data + i00*nb00 + i01*nb01 + i02*nb02 + i03*nb03); + + dst_ptr[id] = *src0_ptr; + id++; + } + } + } + } + } else { + GGML_ASSERT(false); // TODO: implement + } + } + return; + } + // dst counters int64_t i10 = 0; int64_t i11 = 0; @@ -4979,6 +5058,105 @@ static void ggml_compute_forward_dup_f32( return; } + if (src0->type == dst->type && + src0->ne[0] == dst->ne[0] && + src0->nb[0] == GGML_TYPE_SIZE[src0->type] && dst->nb[0] == GGML_TYPE_SIZE[dst->type]) { + // copy by rows + const size_t rs = ne00*nb00; + for (int64_t i03 = 0; i03 < ne03; i03++) { + for (int64_t i02 = 0; i02 < ne02; i02++) { + for (int64_t i01 = 0; i01 < ne01; i01++) { + memcpy( + ((char *) dst->data + i01*nb1 + i02*nb2 + i03*nb3), + ((char *) src0->data + i01*nb01 + i02*nb02 + i03*nb03), + rs); + } + } + } + return; + } + + if (ggml_is_contiguous(dst)) { + // TODO: simplify + if (src0->nb[0] == sizeof(float)) { + if (dst->type == GGML_TYPE_F32) { + size_t id = 0; + const size_t rs = ne00*nb00; + + for (int i03 = 0; i03 < ne03; i03++) { + for (int i02 = 0; i02 < ne02; i02++) { + for (int i01 = 0; i01 < ne01; i01++) { + const char * src0_ptr = (char *) src0->data + i01*nb01 + i02*nb02 + i03*nb03; + char * dst_ptr = (char *) dst->data + id*rs; + + memcpy(dst_ptr, src0_ptr, rs); + + id++; + } + } + } + } else if (dst->type == GGML_TYPE_F16) { + size_t id = 0; + ggml_fp16_t * dst_ptr = (ggml_fp16_t *) dst->data; + + for (int i03 = 0; i03 < ne03; i03++) { + for (int i02 = 0; i02 < ne02; i02++) { + for (int i01 = 0; i01 < ne01; i01++) { + for (int i00 = 0; i00 < ne00; i00++) { + const float * src0_ptr = (float *) ((char *) src0->data + i00*nb00 + i01*nb01 + i02*nb02 + i03*nb03); + + dst_ptr[id] = GGML_FP32_TO_FP16(*src0_ptr); + id++; + } + } + } + } + } else { + GGML_ASSERT(false); // TODO: implement + } + } else { + //printf("%s: this is not optimal - fix me\n", __func__); + + if (dst->type == GGML_TYPE_F32) { + size_t id = 0; + float * dst_ptr = (float *) dst->data; + + for (int i03 = 0; i03 < ne03; i03++) { + for (int i02 = 0; i02 < ne02; i02++) { + for (int i01 = 0; i01 < ne01; i01++) { + for (int i00 = 0; i00 < ne00; i00++) { + const float * src0_ptr = (float *) ((char *) src0->data + i00*nb00 + i01*nb01 + i02*nb02 + i03*nb03); + + dst_ptr[id] = *src0_ptr; + id++; + } + } + } + } + } else if (dst->type == GGML_TYPE_F16) { + size_t id = 0; + ggml_fp16_t * dst_ptr = (ggml_fp16_t *) dst->data; + + for (int i03 = 0; i03 < ne03; i03++) { + for (int i02 = 0; i02 < ne02; i02++) { + for (int i01 = 0; i01 < ne01; i01++) { + for (int i00 = 0; i00 < ne00; i00++) { + const float * src0_ptr = (float *) ((char *) src0->data + i00*nb00 + i01*nb01 + i02*nb02 + i03*nb03); + + dst_ptr[id] = GGML_FP32_TO_FP16(*src0_ptr); + id++; + } + } + } + } + } else { + GGML_ASSERT(false); // TODO: implement + } + } + + return; + } + // dst counters int64_t i10 = 0; int64_t i11 = 0; @@ -5099,14 +5277,18 @@ static void ggml_compute_forward_add_f32( GGML_ASSERT(nb00 == sizeof(float)); if (nb10 == sizeof(float)) { - const int j0 = (n/nth)*ith; - const int j1 = ith == nth - 1 ? n : (n/nth)*(ith + 1); - - for (int j = j0; j < j1; j++) { + for (int j = ith; j < n; j += nth) { +#ifdef GGML_USE_ACCELERATE + vDSP_vadd( + (float *) ((char *) src0->data + j*nb01), 1, + (float *) ((char *) src1->data + j*nb11), 1, + (float *) ((char *) dst->data + j*nb1), 1, nc); +#else ggml_vec_add_f32(nc, (float *) ((char *) dst->data + j*nb1), (float *) ((char *) src0->data + j*nb01), (float *) ((char *) src1->data + j*nb11)); +#endif } } else { // src1 is not contiguous