be/src/storage/index/snii/encoding/crc32c.h
Line | Count | Source |
1 | | // Licensed to the Apache Software Foundation (ASF) under one |
2 | | // or more contributor license agreements. See the NOTICE file |
3 | | // distributed with this work for additional information |
4 | | // regarding copyright ownership. The ASF licenses this file |
5 | | // to you under the Apache License, Version 2.0 (the |
6 | | // "License"); you may not use this file except in compliance |
7 | | // with the License. You may obtain a copy of the License at |
8 | | // |
9 | | // http://www.apache.org/licenses/LICENSE-2.0 |
10 | | // |
11 | | // Unless required by applicable law or agreed to in writing, |
12 | | // software distributed under the License is distributed on an |
13 | | // "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY |
14 | | // KIND, either express or implied. See the License for the |
15 | | // specific language governing permissions and limitations |
16 | | // under the License. |
17 | | |
18 | | #pragma once |
19 | | |
20 | | #include <crc32c/crc32c.h> |
21 | | |
22 | | #include <cstddef> |
23 | | #include <cstdint> |
24 | | |
25 | | #include "storage/index/snii/common/slice.h" |
26 | | |
27 | | namespace doris::snii { |
28 | | |
29 | | // CRC32C (Castagnoli, polynomial 0x1EDC6F41). Used to checksum the tail of each |
30 | | // format block. Thin inline adapter over Doris's bundled Google crc32c thirdparty |
31 | | // (crc32c::Extend / crc32c::Crc32c). That library computes the same canonical |
32 | | // CRC32C (same reflected polynomial, same standard pre/post inversion), so every |
33 | | // on-disk checksum stays byte-identical to the previous in-tree slice-by-8 / |
34 | | // SSE4.2 implementation -- this is an implementation swap, not a format change. |
35 | | // The leading :: keeps the crc32c namespace distinct from crc32c() below. |
36 | 5.60k | inline uint32_t crc32c_extend(uint32_t crc, Slice data) { |
37 | 5.60k | return ::crc32c::Extend(crc, data.data(), data.size()); |
38 | 5.60k | } |
39 | | |
40 | 206k | inline uint32_t crc32c(Slice data) { |
41 | 206k | return ::crc32c::Crc32c(data.data(), data.size()); |
42 | 206k | } |
43 | | |
44 | | #ifdef BE_TEST |
45 | | // T21 test seam. The production crc32c()/crc32c_extend() above delegate to the |
46 | | // bundled Google crc32c thirdparty (see commit d0416bb4129), which already runs a |
47 | | // runtime-dispatched, hardware-accelerated and interleaved CRC32C -- so T21's |
48 | | // "hardware interleaved CRC" goal is already met (and exceeded: that library adds |
49 | | // a PCLMULQDQ fold a hand-rolled 3-way _mm_crc32_u64 lacks). Rather than regress |
50 | | // that reuse, the reference sub-paths below let unit tests prove, byte-for-byte |
51 | | // across all sizes/alignments, that the production path equals the canonical |
52 | | // CRC32C and that the hardware path is engaged: |
53 | | // * crc32c_slice8_extend -- portable software slice-by-8 (always available); |
54 | | // * crc32c_hw_serial_extend -- serial SSE4.2 _mm_crc32 hardware path; |
55 | | // * crc32c_hw3_extend -- 3-way interleaved SSE4.2 hardware path with a |
56 | | // GF(2) shift-combine and a 1024-byte fall-back to |
57 | | // the serial path (the algorithm T21 specifies). |
58 | | // hw_serial/hw3 fall back to slice8 when SSE4.2 is absent. Each *_extend applies |
59 | | // the standard ~crc pre/post inversion, so *_extend(0, d) == crc32c(d). The whole |
60 | | // seam plus its static slice-by-8 table and startup CPUID probe are compiled out |
61 | | // of release builds by this BE_TEST gate, so production carries no extra code. |
62 | | // Pure functions with no shared mutable state (CONCURRENCY: N/A). |
63 | | namespace detail { |
64 | | uint32_t crc32c_slice8_extend(uint32_t crc, Slice data); |
65 | | uint32_t crc32c_hw_serial_extend(uint32_t crc, Slice data); |
66 | | uint32_t crc32c_hw3_extend(uint32_t crc, Slice data); |
67 | | size_t crc32c_interleave_threshold(); |
68 | | bool crc32c_has_hw(); |
69 | | } // namespace detail |
70 | | #endif // BE_TEST |
71 | | |
72 | | } // namespace doris::snii |