[Zlib-devel] pigz 2.1.4 available

[Michael Han]

Mark,

Besides threads, the process() also utilizes many POSIX calls to
iterate directories and files which WINDOWS doesn't support.
It'll be a good idea to abstract IO operations, as yarn did for
threads. Moreover, the IO operation can left to caller by making pigz
buffer driven as what zlib did.. well as you said it's experimental I
think these portability requirements may not make much sense if you
decide to apply the parallel codes to zlib which should be portable by
that time:)

~Michael
On Tue, Nov 11, 2008 at 3:15 PM, Mark Adler <madler at alumni.caltech.edu> wrote:
> On Nov 10, 2008, at 10:45 PM, Michael Han wrote:
>>
>> Essentially the same with what ZLIB is using, just 8 way look up
>> instead of 4.
>
> I will consider adding a BYEIGHT section of code to crc32.c.
>
>> However, the longest_match, which turns out a hot spot during
>> compression, is left intact.
>
> Darn.  That's really the only code that matters for the speed of
> compression.
>
>> Also take a look at the deflate implementation in 7z library.
>
> That's on my to-do list, but it's not high priority.
>
> On Nov 10, 2008, at 10:51 PM, Michael Han wrote:
>>
>> Is there any short term plan to port pigz to windows?
>
> I don't have a windows system, so I can't do it.  All of the pthread calls
> are isolated in yarn.c for just this reason.  All someone would need to do
> is write a different yarn.c that uses windows' native thread functions.
>
> Mark
>
>
> _______________________________________________
> Zlib-devel mailing list
> Zlib-devel at madler.net
> http://mail.madler.net/mailman/listinfo/zlib-devel_madler.net
>



-- 
If first the idea is not absurd, then
there is no hope for it