[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Re: GNU Coding Standards, automake, and the recent xz-utils backdoor
From: |
wrotycz |
Subject: |
Re: Re: GNU Coding Standards, automake, and the recent xz-utils backdoor |
Date: |
Fri, 05 Apr 2024 00:03:51 +0200 |
User-agent: |
GWP-Draft |
> lz4 and zstd are quite a bit less resource-hungry than xz Is it? By
default, and up to -19 level, zstd uses 8 MiB (sliding) window, when xz uses 8
MiB at level 6, which is default, default and doubles is every next level. More
over that, xz compression with 8 MiB window uses 96 MiB of memory, when zstd
needs 272 MiB at level 19, 224 MiB at lv. 16 and 232 MiB at lv. 12. Similarily
with decompression - xz needs 11 MiB to decompress archive with 8 MiB window,
zstd needs 30 MiB to decompress it. I don't see where zstd is 'less
resource-hungry than xz'. > With these options, the zst tarball came
withing a hare's breath of the xz compressed file size. Here are some
samples of about 8 MB files compressed with bzip2, gzip, xz and zstd with
different levels of compression: paste.ee paste.ee/p/tVIWG paste.ee
paste.ee/p/PsNrx > I did not find any drawbacks. Except higher memory
requirements, lower compression, similar issues with long term archiving
usefulness, 'there are no drawbacks'.
- Re: GNU Coding Standards, automake, and the recent xz-utils backdoor, Eric Blake, 2024/04/02
- Re: GNU Coding Standards, automake, and the recent xz-utils backdoor, Karl Berry, 2024/04/02
- Re: GNU Coding Standards, automake, and the recent xz-utils backdoor, Bob Friesenhahn, 2024/04/02
- Re: compressed release distribution formats (was: GNU Coding Standards, automake, and the recent xz-utils backdoor), Jacob Bachmeyer, 2024/04/03
- Re: GNU Coding Standards, automake, and the recent xz-utils backdoor, Richard Stallman, 2024/04/03
- Re: Re: GNU Coding Standards, automake, and the recent xz-utils backdoor,
wrotycz <=
- Re: GNU Coding Standards, automake, and the recent xz-utils backdoor, Richard Stallman, 2024/04/03