[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Daniel Shahaf: Declaration of intent



On Sun, Sep 25, 2016 at 03:14:41PM -0000, Daniel Shahaf wrote:
> I would like to apply to change my status in Debian to Debian
> Maintainer.  I have worked on zsh-syntax-highlighting and reproducible
> builds for a while and I would like to be able to contribute more
> efficiently.
 
\o/ & yay!

while I have never directly worked with Daniel as that I have looked his
work on maintaining packages, I have seen him examing quite many
reproducible builds issues in other peoples packages :-) He's also
active, friendly & competent on the #debian-reproducible channel - so I
very much welcome him to become even more involved in Debian by means of
becoming a Debian maintainer.

a recent highlight of him, his analysis why ksh is unreproducible:

(from https://reproducible.debian.net/ksh - quoted here so it's preserved
once the issue is fixed)

------
Hardcoded values of CHILD_MAX and OPEN_MAX differ.
.
The values for CHILD_MAX are 193122 v. 193527.
.
The values for OPEN_MAX are 16384 v. 65536 (both are powers of two).
.
These are getconf(1) symbolic constants, also defined in the system
header
files, but at least in the CHILD_MAX case I think the difference is
because
src/lib/libast/comp/conf.tab *measures* the constant's value by calling
fork(2) in a loop until that fails. [This can't be the whole
explanation:
the measurement code always prints an even value, but 193527 is odd.]
.
conf.tab has similar code for measuring OPEN_MAX.
----

Lovely, or? 8-)


-- 
cheers,
	Holger

Attachment: signature.asc
Description: Digital signature


Reply to: