作者: 谢瑞阳 10225101483 徐翔宇 10225101535
Du kan inte välja fler än 25 ämnen Ämnen måste starta med en bokstav eller siffra, kan innehålla bindestreck ('-') och vara max 35 tecken långa.

153 rader
3.8 KiB

Replace SSE-optimized CRC32C in POSIX port with external library. Maintaining a hardware-accelerated CRC32C implementation tailored for all modern platforms deserves a repository of its own. We extracted the implementation here into https://github.com/google/crc32c and improved it in that repository. This CL removes the SSE-optimized implementation from this codebase, and adds the ability to use the google/crc32c library, if it is present on the system. The benchmarks below show the performance impact of the change. In summary, open source builds that use the google/crc32c library can expect a 3x improvement in CRC32C throughput, whereas builds that do not use the library will see a 50% drop in CRC32C throughput. This translates in much smaller changes in overall leveldb performance. Baseline, MacBookPro13,3 with Core i7 6920HQ: LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 3.064 micros/op; 36.1 MB/s fillsync : 57.861 micros/op; 1.9 MB/s (1000 ops) fillrandom : 3.887 micros/op; 28.5 MB/s overwrite : 4.140 micros/op; 26.7 MB/s readrandom : 7.433 micros/op; (1000000 of 1000000 found) readrandom : 6.825 micros/op; (1000000 of 1000000 found) readseq : 0.244 micros/op; 453.4 MB/s readreverse : 0.387 micros/op; 285.8 MB/s compact : 449707.000 micros/op; readrandom : 4.196 micros/op; (1000000 of 1000000 found) readseq : 0.228 micros/op; 485.8 MB/s readreverse : 0.320 micros/op; 345.2 MB/s fill100K : 562.556 micros/op; 169.6 MB/s (1000 ops) crc32c : 0.768 micros/op; 5085.0 MB/s (4K per op) snappycomp : 4.220 micros/op; 925.7 MB/s (output: 55.1%) snappyuncomp : 0.635 micros/op; 6155.7 MB/s acquireload : 13.054 micros/op; (each op is 1000 loads) New with crc32c, MacBookPro13,3 with Core i7 6920HQ: LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 2.820 micros/op; 39.2 MB/s fillsync : 51.988 micros/op; 2.1 MB/s (1000 ops) fillrandom : 3.747 micros/op; 29.5 MB/s overwrite : 4.047 micros/op; 27.3 MB/s readrandom : 7.287 micros/op; (1000000 of 1000000 found) readrandom : 6.927 micros/op; (1000000 of 1000000 found) readseq : 0.253 micros/op; 437.5 MB/s readreverse : 0.411 micros/op; 269.2 MB/s compact : 440405.000 micros/op; readrandom : 4.159 micros/op; (1000000 of 1000000 found) readseq : 0.230 micros/op; 481.1 MB/s readreverse : 0.320 micros/op; 345.9 MB/s fill100K : 558.222 micros/op; 170.9 MB/s (1000 ops) crc32c : 0.214 micros/op; 18263.5 MB/s (4K per op) snappycomp : 4.471 micros/op; 873.7 MB/s (output: 55.1%) snappyuncomp : 0.833 micros/op; 4688.5 MB/s acquireload : 13.289 micros/op; (each op is 1000 loads) New without crc32c, MacBookPro13,3 with Core i7 6920HQ LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 3.094 micros/op; 35.8 MB/s fillsync : 52.160 micros/op; 2.1 MB/s (1000 ops) fillrandom : 4.090 micros/op; 27.0 MB/s overwrite : 4.006 micros/op; 27.6 MB/s readrandom : 6.584 micros/op; (1000000 of 1000000 found) readrandom : 6.676 micros/op; (1000000 of 1000000 found) readseq : 0.280 micros/op; 395.2 MB/s readreverse : 0.391 micros/op; 283.2 MB/s compact : 433911.000 micros/op; readrandom : 4.261 micros/op; (1000000 of 1000000 found) readseq : 0.251 micros/op; 440.5 MB/s readreverse : 0.356 micros/op; 310.9 MB/s fill100K : 584.023 micros/op; 163.3 MB/s (1000 ops) crc32c : 1.384 micros/op; 2822.3 MB/s (4K per op) snappycomp : 4.763 micros/op; 820.1 MB/s (output: 55.1%) snappyuncomp : 0.766 micros/op; 5098.6 MB/s acquireload : 12.931 micros/op; (each op is 1000 loads) ------------- Created by MOE: https://github.com/google/moe MOE_MIGRATED_REVID=171667771
7 år sedan
Replace SSE-optimized CRC32C in POSIX port with external library. Maintaining a hardware-accelerated CRC32C implementation tailored for all modern platforms deserves a repository of its own. We extracted the implementation here into https://github.com/google/crc32c and improved it in that repository. This CL removes the SSE-optimized implementation from this codebase, and adds the ability to use the google/crc32c library, if it is present on the system. The benchmarks below show the performance impact of the change. In summary, open source builds that use the google/crc32c library can expect a 3x improvement in CRC32C throughput, whereas builds that do not use the library will see a 50% drop in CRC32C throughput. This translates in much smaller changes in overall leveldb performance. Baseline, MacBookPro13,3 with Core i7 6920HQ: LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 3.064 micros/op; 36.1 MB/s fillsync : 57.861 micros/op; 1.9 MB/s (1000 ops) fillrandom : 3.887 micros/op; 28.5 MB/s overwrite : 4.140 micros/op; 26.7 MB/s readrandom : 7.433 micros/op; (1000000 of 1000000 found) readrandom : 6.825 micros/op; (1000000 of 1000000 found) readseq : 0.244 micros/op; 453.4 MB/s readreverse : 0.387 micros/op; 285.8 MB/s compact : 449707.000 micros/op; readrandom : 4.196 micros/op; (1000000 of 1000000 found) readseq : 0.228 micros/op; 485.8 MB/s readreverse : 0.320 micros/op; 345.2 MB/s fill100K : 562.556 micros/op; 169.6 MB/s (1000 ops) crc32c : 0.768 micros/op; 5085.0 MB/s (4K per op) snappycomp : 4.220 micros/op; 925.7 MB/s (output: 55.1%) snappyuncomp : 0.635 micros/op; 6155.7 MB/s acquireload : 13.054 micros/op; (each op is 1000 loads) New with crc32c, MacBookPro13,3 with Core i7 6920HQ: LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 2.820 micros/op; 39.2 MB/s fillsync : 51.988 micros/op; 2.1 MB/s (1000 ops) fillrandom : 3.747 micros/op; 29.5 MB/s overwrite : 4.047 micros/op; 27.3 MB/s readrandom : 7.287 micros/op; (1000000 of 1000000 found) readrandom : 6.927 micros/op; (1000000 of 1000000 found) readseq : 0.253 micros/op; 437.5 MB/s readreverse : 0.411 micros/op; 269.2 MB/s compact : 440405.000 micros/op; readrandom : 4.159 micros/op; (1000000 of 1000000 found) readseq : 0.230 micros/op; 481.1 MB/s readreverse : 0.320 micros/op; 345.9 MB/s fill100K : 558.222 micros/op; 170.9 MB/s (1000 ops) crc32c : 0.214 micros/op; 18263.5 MB/s (4K per op) snappycomp : 4.471 micros/op; 873.7 MB/s (output: 55.1%) snappyuncomp : 0.833 micros/op; 4688.5 MB/s acquireload : 13.289 micros/op; (each op is 1000 loads) New without crc32c, MacBookPro13,3 with Core i7 6920HQ LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 3.094 micros/op; 35.8 MB/s fillsync : 52.160 micros/op; 2.1 MB/s (1000 ops) fillrandom : 4.090 micros/op; 27.0 MB/s overwrite : 4.006 micros/op; 27.6 MB/s readrandom : 6.584 micros/op; (1000000 of 1000000 found) readrandom : 6.676 micros/op; (1000000 of 1000000 found) readseq : 0.280 micros/op; 395.2 MB/s readreverse : 0.391 micros/op; 283.2 MB/s compact : 433911.000 micros/op; readrandom : 4.261 micros/op; (1000000 of 1000000 found) readseq : 0.251 micros/op; 440.5 MB/s readreverse : 0.356 micros/op; 310.9 MB/s fill100K : 584.023 micros/op; 163.3 MB/s (1000 ops) crc32c : 1.384 micros/op; 2822.3 MB/s (4K per op) snappycomp : 4.763 micros/op; 820.1 MB/s (output: 55.1%) snappyuncomp : 0.766 micros/op; 5098.6 MB/s acquireload : 12.931 micros/op; (each op is 1000 loads) ------------- Created by MOE: https://github.com/google/moe MOE_MIGRATED_REVID=171667771
7 år sedan
Replace SSE-optimized CRC32C in POSIX port with external library. Maintaining a hardware-accelerated CRC32C implementation tailored for all modern platforms deserves a repository of its own. We extracted the implementation here into https://github.com/google/crc32c and improved it in that repository. This CL removes the SSE-optimized implementation from this codebase, and adds the ability to use the google/crc32c library, if it is present on the system. The benchmarks below show the performance impact of the change. In summary, open source builds that use the google/crc32c library can expect a 3x improvement in CRC32C throughput, whereas builds that do not use the library will see a 50% drop in CRC32C throughput. This translates in much smaller changes in overall leveldb performance. Baseline, MacBookPro13,3 with Core i7 6920HQ: LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 3.064 micros/op; 36.1 MB/s fillsync : 57.861 micros/op; 1.9 MB/s (1000 ops) fillrandom : 3.887 micros/op; 28.5 MB/s overwrite : 4.140 micros/op; 26.7 MB/s readrandom : 7.433 micros/op; (1000000 of 1000000 found) readrandom : 6.825 micros/op; (1000000 of 1000000 found) readseq : 0.244 micros/op; 453.4 MB/s readreverse : 0.387 micros/op; 285.8 MB/s compact : 449707.000 micros/op; readrandom : 4.196 micros/op; (1000000 of 1000000 found) readseq : 0.228 micros/op; 485.8 MB/s readreverse : 0.320 micros/op; 345.2 MB/s fill100K : 562.556 micros/op; 169.6 MB/s (1000 ops) crc32c : 0.768 micros/op; 5085.0 MB/s (4K per op) snappycomp : 4.220 micros/op; 925.7 MB/s (output: 55.1%) snappyuncomp : 0.635 micros/op; 6155.7 MB/s acquireload : 13.054 micros/op; (each op is 1000 loads) New with crc32c, MacBookPro13,3 with Core i7 6920HQ: LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 2.820 micros/op; 39.2 MB/s fillsync : 51.988 micros/op; 2.1 MB/s (1000 ops) fillrandom : 3.747 micros/op; 29.5 MB/s overwrite : 4.047 micros/op; 27.3 MB/s readrandom : 7.287 micros/op; (1000000 of 1000000 found) readrandom : 6.927 micros/op; (1000000 of 1000000 found) readseq : 0.253 micros/op; 437.5 MB/s readreverse : 0.411 micros/op; 269.2 MB/s compact : 440405.000 micros/op; readrandom : 4.159 micros/op; (1000000 of 1000000 found) readseq : 0.230 micros/op; 481.1 MB/s readreverse : 0.320 micros/op; 345.9 MB/s fill100K : 558.222 micros/op; 170.9 MB/s (1000 ops) crc32c : 0.214 micros/op; 18263.5 MB/s (4K per op) snappycomp : 4.471 micros/op; 873.7 MB/s (output: 55.1%) snappyuncomp : 0.833 micros/op; 4688.5 MB/s acquireload : 13.289 micros/op; (each op is 1000 loads) New without crc32c, MacBookPro13,3 with Core i7 6920HQ LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 3.094 micros/op; 35.8 MB/s fillsync : 52.160 micros/op; 2.1 MB/s (1000 ops) fillrandom : 4.090 micros/op; 27.0 MB/s overwrite : 4.006 micros/op; 27.6 MB/s readrandom : 6.584 micros/op; (1000000 of 1000000 found) readrandom : 6.676 micros/op; (1000000 of 1000000 found) readseq : 0.280 micros/op; 395.2 MB/s readreverse : 0.391 micros/op; 283.2 MB/s compact : 433911.000 micros/op; readrandom : 4.261 micros/op; (1000000 of 1000000 found) readseq : 0.251 micros/op; 440.5 MB/s readreverse : 0.356 micros/op; 310.9 MB/s fill100K : 584.023 micros/op; 163.3 MB/s (1000 ops) crc32c : 1.384 micros/op; 2822.3 MB/s (4K per op) snappycomp : 4.763 micros/op; 820.1 MB/s (output: 55.1%) snappyuncomp : 0.766 micros/op; 5098.6 MB/s acquireload : 12.931 micros/op; (each op is 1000 loads) ------------- Created by MOE: https://github.com/google/moe MOE_MIGRATED_REVID=171667771
7 år sedan
Replace SSE-optimized CRC32C in POSIX port with external library. Maintaining a hardware-accelerated CRC32C implementation tailored for all modern platforms deserves a repository of its own. We extracted the implementation here into https://github.com/google/crc32c and improved it in that repository. This CL removes the SSE-optimized implementation from this codebase, and adds the ability to use the google/crc32c library, if it is present on the system. The benchmarks below show the performance impact of the change. In summary, open source builds that use the google/crc32c library can expect a 3x improvement in CRC32C throughput, whereas builds that do not use the library will see a 50% drop in CRC32C throughput. This translates in much smaller changes in overall leveldb performance. Baseline, MacBookPro13,3 with Core i7 6920HQ: LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 3.064 micros/op; 36.1 MB/s fillsync : 57.861 micros/op; 1.9 MB/s (1000 ops) fillrandom : 3.887 micros/op; 28.5 MB/s overwrite : 4.140 micros/op; 26.7 MB/s readrandom : 7.433 micros/op; (1000000 of 1000000 found) readrandom : 6.825 micros/op; (1000000 of 1000000 found) readseq : 0.244 micros/op; 453.4 MB/s readreverse : 0.387 micros/op; 285.8 MB/s compact : 449707.000 micros/op; readrandom : 4.196 micros/op; (1000000 of 1000000 found) readseq : 0.228 micros/op; 485.8 MB/s readreverse : 0.320 micros/op; 345.2 MB/s fill100K : 562.556 micros/op; 169.6 MB/s (1000 ops) crc32c : 0.768 micros/op; 5085.0 MB/s (4K per op) snappycomp : 4.220 micros/op; 925.7 MB/s (output: 55.1%) snappyuncomp : 0.635 micros/op; 6155.7 MB/s acquireload : 13.054 micros/op; (each op is 1000 loads) New with crc32c, MacBookPro13,3 with Core i7 6920HQ: LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 2.820 micros/op; 39.2 MB/s fillsync : 51.988 micros/op; 2.1 MB/s (1000 ops) fillrandom : 3.747 micros/op; 29.5 MB/s overwrite : 4.047 micros/op; 27.3 MB/s readrandom : 7.287 micros/op; (1000000 of 1000000 found) readrandom : 6.927 micros/op; (1000000 of 1000000 found) readseq : 0.253 micros/op; 437.5 MB/s readreverse : 0.411 micros/op; 269.2 MB/s compact : 440405.000 micros/op; readrandom : 4.159 micros/op; (1000000 of 1000000 found) readseq : 0.230 micros/op; 481.1 MB/s readreverse : 0.320 micros/op; 345.9 MB/s fill100K : 558.222 micros/op; 170.9 MB/s (1000 ops) crc32c : 0.214 micros/op; 18263.5 MB/s (4K per op) snappycomp : 4.471 micros/op; 873.7 MB/s (output: 55.1%) snappyuncomp : 0.833 micros/op; 4688.5 MB/s acquireload : 13.289 micros/op; (each op is 1000 loads) New without crc32c, MacBookPro13,3 with Core i7 6920HQ LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 3.094 micros/op; 35.8 MB/s fillsync : 52.160 micros/op; 2.1 MB/s (1000 ops) fillrandom : 4.090 micros/op; 27.0 MB/s overwrite : 4.006 micros/op; 27.6 MB/s readrandom : 6.584 micros/op; (1000000 of 1000000 found) readrandom : 6.676 micros/op; (1000000 of 1000000 found) readseq : 0.280 micros/op; 395.2 MB/s readreverse : 0.391 micros/op; 283.2 MB/s compact : 433911.000 micros/op; readrandom : 4.261 micros/op; (1000000 of 1000000 found) readseq : 0.251 micros/op; 440.5 MB/s readreverse : 0.356 micros/op; 310.9 MB/s fill100K : 584.023 micros/op; 163.3 MB/s (1000 ops) crc32c : 1.384 micros/op; 2822.3 MB/s (4K per op) snappycomp : 4.763 micros/op; 820.1 MB/s (output: 55.1%) snappyuncomp : 0.766 micros/op; 5098.6 MB/s acquireload : 12.931 micros/op; (each op is 1000 loads) ------------- Created by MOE: https://github.com/google/moe MOE_MIGRATED_REVID=171667771
7 år sedan
Replace SSE-optimized CRC32C in POSIX port with external library. Maintaining a hardware-accelerated CRC32C implementation tailored for all modern platforms deserves a repository of its own. We extracted the implementation here into https://github.com/google/crc32c and improved it in that repository. This CL removes the SSE-optimized implementation from this codebase, and adds the ability to use the google/crc32c library, if it is present on the system. The benchmarks below show the performance impact of the change. In summary, open source builds that use the google/crc32c library can expect a 3x improvement in CRC32C throughput, whereas builds that do not use the library will see a 50% drop in CRC32C throughput. This translates in much smaller changes in overall leveldb performance. Baseline, MacBookPro13,3 with Core i7 6920HQ: LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 3.064 micros/op; 36.1 MB/s fillsync : 57.861 micros/op; 1.9 MB/s (1000 ops) fillrandom : 3.887 micros/op; 28.5 MB/s overwrite : 4.140 micros/op; 26.7 MB/s readrandom : 7.433 micros/op; (1000000 of 1000000 found) readrandom : 6.825 micros/op; (1000000 of 1000000 found) readseq : 0.244 micros/op; 453.4 MB/s readreverse : 0.387 micros/op; 285.8 MB/s compact : 449707.000 micros/op; readrandom : 4.196 micros/op; (1000000 of 1000000 found) readseq : 0.228 micros/op; 485.8 MB/s readreverse : 0.320 micros/op; 345.2 MB/s fill100K : 562.556 micros/op; 169.6 MB/s (1000 ops) crc32c : 0.768 micros/op; 5085.0 MB/s (4K per op) snappycomp : 4.220 micros/op; 925.7 MB/s (output: 55.1%) snappyuncomp : 0.635 micros/op; 6155.7 MB/s acquireload : 13.054 micros/op; (each op is 1000 loads) New with crc32c, MacBookPro13,3 with Core i7 6920HQ: LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 2.820 micros/op; 39.2 MB/s fillsync : 51.988 micros/op; 2.1 MB/s (1000 ops) fillrandom : 3.747 micros/op; 29.5 MB/s overwrite : 4.047 micros/op; 27.3 MB/s readrandom : 7.287 micros/op; (1000000 of 1000000 found) readrandom : 6.927 micros/op; (1000000 of 1000000 found) readseq : 0.253 micros/op; 437.5 MB/s readreverse : 0.411 micros/op; 269.2 MB/s compact : 440405.000 micros/op; readrandom : 4.159 micros/op; (1000000 of 1000000 found) readseq : 0.230 micros/op; 481.1 MB/s readreverse : 0.320 micros/op; 345.9 MB/s fill100K : 558.222 micros/op; 170.9 MB/s (1000 ops) crc32c : 0.214 micros/op; 18263.5 MB/s (4K per op) snappycomp : 4.471 micros/op; 873.7 MB/s (output: 55.1%) snappyuncomp : 0.833 micros/op; 4688.5 MB/s acquireload : 13.289 micros/op; (each op is 1000 loads) New without crc32c, MacBookPro13,3 with Core i7 6920HQ LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 3.094 micros/op; 35.8 MB/s fillsync : 52.160 micros/op; 2.1 MB/s (1000 ops) fillrandom : 4.090 micros/op; 27.0 MB/s overwrite : 4.006 micros/op; 27.6 MB/s readrandom : 6.584 micros/op; (1000000 of 1000000 found) readrandom : 6.676 micros/op; (1000000 of 1000000 found) readseq : 0.280 micros/op; 395.2 MB/s readreverse : 0.391 micros/op; 283.2 MB/s compact : 433911.000 micros/op; readrandom : 4.261 micros/op; (1000000 of 1000000 found) readseq : 0.251 micros/op; 440.5 MB/s readreverse : 0.356 micros/op; 310.9 MB/s fill100K : 584.023 micros/op; 163.3 MB/s (1000 ops) crc32c : 1.384 micros/op; 2822.3 MB/s (4K per op) snappycomp : 4.763 micros/op; 820.1 MB/s (output: 55.1%) snappyuncomp : 0.766 micros/op; 5098.6 MB/s acquireload : 12.931 micros/op; (each op is 1000 loads) ------------- Created by MOE: https://github.com/google/moe MOE_MIGRATED_REVID=171667771
7 år sedan
Replace SSE-optimized CRC32C in POSIX port with external library. Maintaining a hardware-accelerated CRC32C implementation tailored for all modern platforms deserves a repository of its own. We extracted the implementation here into https://github.com/google/crc32c and improved it in that repository. This CL removes the SSE-optimized implementation from this codebase, and adds the ability to use the google/crc32c library, if it is present on the system. The benchmarks below show the performance impact of the change. In summary, open source builds that use the google/crc32c library can expect a 3x improvement in CRC32C throughput, whereas builds that do not use the library will see a 50% drop in CRC32C throughput. This translates in much smaller changes in overall leveldb performance. Baseline, MacBookPro13,3 with Core i7 6920HQ: LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 3.064 micros/op; 36.1 MB/s fillsync : 57.861 micros/op; 1.9 MB/s (1000 ops) fillrandom : 3.887 micros/op; 28.5 MB/s overwrite : 4.140 micros/op; 26.7 MB/s readrandom : 7.433 micros/op; (1000000 of 1000000 found) readrandom : 6.825 micros/op; (1000000 of 1000000 found) readseq : 0.244 micros/op; 453.4 MB/s readreverse : 0.387 micros/op; 285.8 MB/s compact : 449707.000 micros/op; readrandom : 4.196 micros/op; (1000000 of 1000000 found) readseq : 0.228 micros/op; 485.8 MB/s readreverse : 0.320 micros/op; 345.2 MB/s fill100K : 562.556 micros/op; 169.6 MB/s (1000 ops) crc32c : 0.768 micros/op; 5085.0 MB/s (4K per op) snappycomp : 4.220 micros/op; 925.7 MB/s (output: 55.1%) snappyuncomp : 0.635 micros/op; 6155.7 MB/s acquireload : 13.054 micros/op; (each op is 1000 loads) New with crc32c, MacBookPro13,3 with Core i7 6920HQ: LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 2.820 micros/op; 39.2 MB/s fillsync : 51.988 micros/op; 2.1 MB/s (1000 ops) fillrandom : 3.747 micros/op; 29.5 MB/s overwrite : 4.047 micros/op; 27.3 MB/s readrandom : 7.287 micros/op; (1000000 of 1000000 found) readrandom : 6.927 micros/op; (1000000 of 1000000 found) readseq : 0.253 micros/op; 437.5 MB/s readreverse : 0.411 micros/op; 269.2 MB/s compact : 440405.000 micros/op; readrandom : 4.159 micros/op; (1000000 of 1000000 found) readseq : 0.230 micros/op; 481.1 MB/s readreverse : 0.320 micros/op; 345.9 MB/s fill100K : 558.222 micros/op; 170.9 MB/s (1000 ops) crc32c : 0.214 micros/op; 18263.5 MB/s (4K per op) snappycomp : 4.471 micros/op; 873.7 MB/s (output: 55.1%) snappyuncomp : 0.833 micros/op; 4688.5 MB/s acquireload : 13.289 micros/op; (each op is 1000 loads) New without crc32c, MacBookPro13,3 with Core i7 6920HQ LevelDB: version 1.20 Keys: 16 bytes each Values: 100 bytes each (50 bytes after compression) Entries: 1000000 RawSize: 110.6 MB (estimated) FileSize: 62.9 MB (estimated) ------------------------------------------------ fillseq : 3.094 micros/op; 35.8 MB/s fillsync : 52.160 micros/op; 2.1 MB/s (1000 ops) fillrandom : 4.090 micros/op; 27.0 MB/s overwrite : 4.006 micros/op; 27.6 MB/s readrandom : 6.584 micros/op; (1000000 of 1000000 found) readrandom : 6.676 micros/op; (1000000 of 1000000 found) readseq : 0.280 micros/op; 395.2 MB/s readreverse : 0.391 micros/op; 283.2 MB/s compact : 433911.000 micros/op; readrandom : 4.261 micros/op; (1000000 of 1000000 found) readseq : 0.251 micros/op; 440.5 MB/s readreverse : 0.356 micros/op; 310.9 MB/s fill100K : 584.023 micros/op; 163.3 MB/s (1000 ops) crc32c : 1.384 micros/op; 2822.3 MB/s (4K per op) snappycomp : 4.763 micros/op; 820.1 MB/s (output: 55.1%) snappyuncomp : 0.766 micros/op; 5098.6 MB/s acquireload : 12.931 micros/op; (each op is 1000 loads) ------------- Created by MOE: https://github.com/google/moe MOE_MIGRATED_REVID=171667771
7 år sedan
  1. // Copyright (c) 2018 The LevelDB Authors. All rights reserved.
  2. // Use of this source code is governed by a BSD-style license that can be
  3. // found in the LICENSE file. See the AUTHORS file for names of contributors.
  4. #ifndef STORAGE_LEVELDB_PORT_PORT_STDCXX_H_
  5. #define STORAGE_LEVELDB_PORT_PORT_STDCXX_H_
  6. // port/port_config.h availability is automatically detected via __has_include
  7. // in newer compilers. If LEVELDB_HAS_PORT_CONFIG_H is defined, it overrides the
  8. // configuration detection.
  9. #if defined(LEVELDB_HAS_PORT_CONFIG_H)
  10. #if LEVELDB_HAS_PORT_CONFIG_H
  11. #include "port/port_config.h"
  12. #endif // LEVELDB_HAS_PORT_CONFIG_H
  13. #elif defined(__has_include)
  14. #if __has_include("port/port_config.h")
  15. #include "port/port_config.h"
  16. #endif // __has_include("port/port_config.h")
  17. #endif // defined(LEVELDB_HAS_PORT_CONFIG_H)
  18. #if HAVE_CRC32C
  19. #include <crc32c/crc32c.h>
  20. #endif // HAVE_CRC32C
  21. #if HAVE_SNAPPY
  22. #include <snappy.h>
  23. #endif // HAVE_SNAPPY
  24. #include <cassert>
  25. #include <condition_variable> // NOLINT
  26. #include <cstddef>
  27. #include <cstdint>
  28. #include <mutex> // NOLINT
  29. #include <string>
  30. #include "port/thread_annotations.h"
  31. namespace leveldb {
  32. namespace port {
  33. static const bool kLittleEndian = !LEVELDB_IS_BIG_ENDIAN;
  34. class CondVar;
  35. // Thinly wraps std::mutex.
  36. class LOCKABLE Mutex {
  37. public:
  38. Mutex() = default;
  39. ~Mutex() = default;
  40. Mutex(const Mutex&) = delete;
  41. Mutex& operator=(const Mutex&) = delete;
  42. void Lock() EXCLUSIVE_LOCK_FUNCTION() { mu_.lock(); }
  43. void Unlock() UNLOCK_FUNCTION() { mu_.unlock(); }
  44. void AssertHeld() ASSERT_EXCLUSIVE_LOCK() {}
  45. private:
  46. friend class CondVar;
  47. std::mutex mu_;
  48. };
  49. // Thinly wraps std::condition_variable.
  50. class CondVar {
  51. public:
  52. explicit CondVar(Mutex* mu) : mu_(mu) { assert(mu != nullptr); }
  53. ~CondVar() = default;
  54. CondVar(const CondVar&) = delete;
  55. CondVar& operator=(const CondVar&) = delete;
  56. void Wait() {
  57. std::unique_lock<std::mutex> lock(mu_->mu_, std::adopt_lock);
  58. cv_.wait(lock);
  59. lock.release();
  60. }
  61. void Signal() { cv_.notify_one(); }
  62. void SignalAll() { cv_.notify_all(); }
  63. private:
  64. std::condition_variable cv_;
  65. Mutex* const mu_;
  66. };
  67. inline bool Snappy_Compress(const char* input, size_t length,
  68. std::string* output) {
  69. #if HAVE_SNAPPY
  70. output->resize(snappy::MaxCompressedLength(length));
  71. size_t outlen;
  72. snappy::RawCompress(input, length, &(*output)[0], &outlen);
  73. output->resize(outlen);
  74. return true;
  75. #else
  76. // Silence compiler warnings about unused arguments.
  77. (void)input;
  78. (void)length;
  79. (void)output;
  80. #endif // HAVE_SNAPPY
  81. return false;
  82. }
  83. inline bool Snappy_GetUncompressedLength(const char* input, size_t length,
  84. size_t* result) {
  85. #if HAVE_SNAPPY
  86. return snappy::GetUncompressedLength(input, length, result);
  87. #else
  88. // Silence compiler warnings about unused arguments.
  89. (void)input;
  90. (void)length;
  91. (void)result;
  92. return false;
  93. #endif // HAVE_SNAPPY
  94. }
  95. inline bool Snappy_Uncompress(const char* input, size_t length, char* output) {
  96. #if HAVE_SNAPPY
  97. return snappy::RawUncompress(input, length, output);
  98. #else
  99. // Silence compiler warnings about unused arguments.
  100. (void)input;
  101. (void)length;
  102. (void)output;
  103. return false;
  104. #endif // HAVE_SNAPPY
  105. }
  106. inline bool GetHeapProfile(void (*func)(void*, const char*, int), void* arg) {
  107. // Silence compiler warnings about unused arguments.
  108. (void)func;
  109. (void)arg;
  110. return false;
  111. }
  112. inline uint32_t AcceleratedCRC32C(uint32_t crc, const char* buf, size_t size) {
  113. #if HAVE_CRC32C
  114. return ::crc32c::Extend(crc, reinterpret_cast<const uint8_t*>(buf), size);
  115. #else
  116. // Silence compiler warnings about unused arguments.
  117. (void)crc;
  118. (void)buf;
  119. (void)size;
  120. return 0;
  121. #endif // HAVE_CRC32C
  122. }
  123. } // namespace port
  124. } // namespace leveldb
  125. #endif // STORAGE_LEVELDB_PORT_PORT_STDCXX_H_