oxicuda-ptx 0.2.0

OxiCUDA PTX - PTX code generation DSL and IR for GPU kernel development
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138

//! Secondary `ElementwiseTemplate` method implementations.
//!
//! Contains additional `ElementwiseTemplate` inherent methods that did
//! not fit in the primary impl module: `generate_xor`, `generate_fused_scale_add`,
//! and `generate_fill`.
//!
//! Refactored with [SplitRS](https://github.com/cool-japan/splitrs).

use crate::builder::KernelBuilder;
use crate::error::PtxGenError;
use crate::ir::PtxType;

use super::elementwisetemplate_type::ElementwiseTemplate;
use super::functions::{float_two_literal, scalar_param_type};

impl ElementwiseTemplate {
    /// Generates a fuzzy XOR kernel: `c[i] = a[i] + b[i] - 2*a[i]*b[i]`.
    pub(super) fn generate_xor(&self) -> Result<String, PtxGenError> {
        let kernel_name = self.kernel_name();
        let ty = self.ty_str();
        let byte_size = self.precision.size_bytes();
        let two_lit = float_two_literal(self.precision);
        KernelBuilder::new(&kernel_name)
            .target(self.target)
            .param("a_ptr", PtxType::U64)
            .param("b_ptr", PtxType::U64)
            .param("c_ptr", PtxType::U64)
            .param("n", PtxType::U32)
            .max_threads_per_block(256)
            .body(move |b| {
                let tid = b.global_thread_id_x();
                let tid_name = tid.to_string();
                let n_reg = b.load_param_u32("n");
                b.if_lt_u32(tid, n_reg, move |b| {
                    let a_ptr = b.load_param_u64("a_ptr");
                    let b_ptr = b.load_param_u64("b_ptr");
                    let c_ptr = b.load_param_u64("c_ptr");
                    b.raw_ptx(&format!(
                        "cvt.u64.u32 %rd_off, {tid_name};\n    \
                         mul.lo.u64 %rd_off, %rd_off, {byte_size};\n    \
                         add.u64 %rd_a, {a_ptr}, %rd_off;\n    \
                         add.u64 %rd_b, {b_ptr}, %rd_off;\n    \
                         add.u64 %rd_c, {c_ptr}, %rd_off;"
                    ));
                    b.raw_ptx(&format!(
                        "ld.global{ty} %f_a, [%rd_a];\n    \
                         ld.global{ty} %f_b, [%rd_b];\n    \
                         add{ty} %f_s, %f_a, %f_b;\n    \
                         mul{ty} %f_t, %f_a, %f_b;\n    \
                         mul{ty} %f_t2, %f_t, {two_lit};\n    \
                         sub{ty} %f_c, %f_s, %f_t2;\n    \
                         st.global{ty} [%rd_c], %f_c;"
                    ));
                });
                b.ret();
            })
            .build()
    }
    /// Generates a fused scale-add kernel: `c[i] = alpha * a[i] + beta * b[i]`.
    pub(super) fn generate_fused_scale_add(&self) -> Result<String, PtxGenError> {
        let kernel_name = self.kernel_name();
        let ty = self.ty_str();
        let byte_size = self.precision.size_bytes();
        let scalar_ty = scalar_param_type(self.precision);
        KernelBuilder::new(&kernel_name)
            .target(self.target)
            .param("a_ptr", PtxType::U64)
            .param("b_ptr", PtxType::U64)
            .param("c_ptr", PtxType::U64)
            .param("alpha", scalar_ty)
            .param("beta", scalar_ty)
            .param("n", PtxType::U32)
            .max_threads_per_block(256)
            .body(move |b| {
                let tid = b.global_thread_id_x();
                let tid_name = tid.to_string();
                let n_reg = b.load_param_u32("n");
                b.if_lt_u32(tid, n_reg, move |b| {
                    let a_ptr = b.load_param_u64("a_ptr");
                    let b_ptr = b.load_param_u64("b_ptr");
                    let c_ptr = b.load_param_u64("c_ptr");
                    b.raw_ptx(&format!(
                        "cvt.u64.u32 %rd_off, {tid_name};\n    \
                         mul.lo.u64 %rd_off, %rd_off, {byte_size};\n    \
                         add.u64 %rd_a, {a_ptr}, %rd_off;\n    \
                         add.u64 %rd_b, {b_ptr}, %rd_off;\n    \
                         add.u64 %rd_c, {c_ptr}, %rd_off;"
                    ));
                    b.raw_ptx(&format!(
                        "ld.param{ty} %f_alpha, [%param_alpha];\n    \
                         ld.param{ty} %f_beta, [%param_beta];\n    \
                         ld.global{ty} %f_a, [%rd_a];\n    \
                         ld.global{ty} %f_b, [%rd_b];\n    \
                         mul{ty} %f_aa, %f_alpha, %f_a;\n    \
                         mul{ty} %f_bb, %f_beta, %f_b;\n    \
                         add{ty} %f_y, %f_aa, %f_bb;\n    \
                         st.global{ty} [%rd_c], %f_y;"
                    ));
                });
                b.ret();
            })
            .build()
    }
    /// Generates a fill kernel: `dst[i] = value` for all `i < n`.
    ///
    /// The scalar `value` is read from a kernel parameter (not from a source buffer),
    /// which means every output element receives the same constant. This mirrors the
    /// `generate_scale()` pattern for loading a scalar kernel parameter via `ld.param`.
    pub(super) fn generate_fill(&self) -> Result<String, PtxGenError> {
        let kernel_name = self.kernel_name();
        let ty = self.ty_str();
        let byte_size = self.precision.size_bytes();
        let scalar_ty = scalar_param_type(self.precision);
        KernelBuilder::new(&kernel_name)
            .target(self.target)
            .param("dst_ptr", PtxType::U64)
            .param("value", scalar_ty)
            .param("n", PtxType::U32)
            .max_threads_per_block(256)
            .body(move |b| {
                let tid = b.global_thread_id_x();
                let tid_name = tid.to_string();
                let n_reg = b.load_param_u32("n");
                b.if_lt_u32(tid, n_reg, move |b| {
                    let dst_ptr = b.load_param_u64("dst_ptr");
                    b.raw_ptx(&format!(
                        "cvt.u64.u32 %rd_off, {tid_name};\n    \
                         mul.lo.u64 %rd_off, %rd_off, {byte_size};\n    \
                         add.u64 %rd_dst, {dst_ptr}, %rd_off;\n    \
                         ld.param{ty} %f_val, [%param_value];\n    \
                         st.global{ty} [%rd_dst], %f_val;"
                    ));
                });
                b.ret();
            })
            .build()
    }
}