File: SIProgramInfo.h

package info (click to toggle)
llvm-toolchain-9 1%3A9.0.1-16.1
  • links: PTS, VCS
  • area: main
  • in suites: bullseye
  • size: 882,388 kB
  • sloc: cpp: 4,167,636; ansic: 714,256; asm: 457,610; python: 155,927; objc: 65,094; sh: 42,856; lisp: 26,908; perl: 7,786; pascal: 7,722; makefile: 6,881; ml: 5,581; awk: 3,648; cs: 2,027; xml: 888; javascript: 381; ruby: 156
file content (66 lines) | stat: -rw-r--r-- 1,973 bytes parent folder | download | duplicates (2)
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
//===--- SIProgramInfo.h ----------------------------------------*- C++ -*-===//
//
// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
// See https://llvm.org/LICENSE.txt for license information.
// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
//
//===----------------------------------------------------------------------===//
//
/// \file
/// Defines struct to track resource usage for kernels and entry functions.
///
//
//===----------------------------------------------------------------------===//

#ifndef LLVM_LIB_TARGET_AMDGPU_SIPROGRAMINFO_H
#define LLVM_LIB_TARGET_AMDGPU_SIPROGRAMINFO_H

namespace llvm {

/// Track resource usage for kernels / entry functions.
struct SIProgramInfo {
    // Fields set in PGM_RSRC1 pm4 packet.
    uint32_t VGPRBlocks = 0;
    uint32_t SGPRBlocks = 0;
    uint32_t Priority = 0;
    uint32_t FloatMode = 0;
    uint32_t Priv = 0;
    uint32_t DX10Clamp = 0;
    uint32_t DebugMode = 0;
    uint32_t IEEEMode = 0;
    uint32_t WgpMode = 0; // GFX10+
    uint32_t MemOrdered = 0; // GFX10+
    uint64_t ScratchSize = 0;

    uint64_t ComputePGMRSrc1 = 0;

    // Fields set in PGM_RSRC2 pm4 packet.
    uint32_t LDSBlocks = 0;
    uint32_t ScratchBlocks = 0;

    uint64_t ComputePGMRSrc2 = 0;

    uint32_t NumVGPR = 0;
    uint32_t NumSGPR = 0;
    uint32_t LDSSize = 0;
    bool FlatUsed = false;

    // Number of SGPRs that meets number of waves per execution unit request.
    uint32_t NumSGPRsForWavesPerEU = 0;

    // Number of VGPRs that meets number of waves per execution unit request.
    uint32_t NumVGPRsForWavesPerEU = 0;

    // Whether there is recursion, dynamic allocas, indirect calls or some other
    // reason there may be statically unknown stack usage.
    bool DynamicCallStack = false;

    // Bonus information for debugging.
    bool VCCUsed = false;

    SIProgramInfo() = default;
};

} // namespace llvm

#endif // LLVM_LIB_TARGET_AMDGPU_SIPROGRAMINFO_H