Giters

kspalaiologos / bzip3

A better and stronger spiritual successor to BZip2.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Parallel data compression

lskibu opened this issue · comments

commented

The option -j --jobs specifies the number of parallel workers I suppose
but I see no use of threads or forking...
Plus it's not win32/64 compatible as no <windows.h> was included in the source
and specifically in main.c

I like how the linear suffix array library was implemented & hardcoded
but the -j means nothing if no extra CPU cores can be used during compression/decompression

commented

The option -j --jobs specifies the number of parallel workers I suppose

Correct.

but I see no use of threads or forking...

https://github.com/kspalaiologos/bzip3/blob/master/src/main.c#L216
https://github.com/kspalaiologos/bzip3/blob/master/src/main.c#L246
https://github.com/kspalaiologos/bzip3/blob/master/src/libbz3.c#L748

Plus it's not win32/64 compatible as no <windows.h> was included in the source

https://github.com/kspalaiologos/bzip3/blob/master/.github/workflows/release.yml#L58

I like how the linear suffix array library was implemented & hardcoded

Because it was buggy, the upstream maintainer refused to fix it so I had to do it myself: IlyaGrebnov/libsais#10