[llvm] [LLVM] add LZMA for compression/decompression (PR #83297)

Yaxun Liu via llvm-commits llvm-commits at lists.llvm.org
Thu Feb 29 10:41:54 PST 2024


yxsamliu wrote:

The following is measurement for compressing/decompressing Blender 4.1 bundled bitcode for 6 GPU arch's:

It is surprising that LZMA level 9 gets higher compressing rate with less compressing/decompressing time, but it did happen.

| Compression Method | Level | Original Size (bytes) | Compressed Size (bytes) | Compression Rate | Compression Time (s) | Decompression Time (s) |
|---------------------|-------|-----------------------|-------------------------|------------------|----------------------|------------------------|
| LZMA                | 6     | 68,459,756            | 22,984,456              | ~2.98:1          | 18.7226              | 1.1000                 |
| LZMA                | 9     | 68,459,756            | 4,139,504               | ~16.55:1         | 14.32                | 0.3012                 |
| ZSTD                | 6     | 68,459,756            | 32,612,291              | ~2.10:1          | 0.8067               | 0.0982                 |
| ZSTD                | 9     | 68,459,756            | 31,445,373              | ~2.18:1          | 1.3375               | 0.0933                 |


https://github.com/llvm/llvm-project/pull/83297


More information about the llvm-commits mailing list