This is the mail archive of the
mailing list for the Cygwin project.
Re: Safeness of arc4random while fork
- From: Brian Inglis <Brian dot Inglis at SystematicSw dot ab dot ca>
- To: cygwin at cygwin dot com
- Date: Sun, 30 Apr 2017 11:30:23 -0600
- Subject: Re: Safeness of arc4random while fork
- Authentication-results: sourceware.org; auth=none
- References: <CAB2YY7FGEcAyVmCBJZwJ+mm4DzrWxyQbR-a8rw7YdYnR2aRxTg@mail.gmail.com>
- Reply-to: Brian dot Inglis at SystematicSw dot ab dot ca
On 2017-04-30 07:12, Kinichiro Inoguchi wrote:
> I saw this issue while building LibreSSL on Cygwin64.
> I had downloaded LibreSSL tar ball from here.
> And I tried to build it, build was ok, but 1 regression test failed.
> This test is testing the safeness of arc4random() while forking.
> With other platforms, linux, *bsd, OSX, MinGW64, this test never
> I checked FAQ and ML archives, but couldn't find anything.
> Has anyone experienced this already ?
> Could someone check the implementation of arc4random functions,
> please ?
> FAIL: arc4randomforktest.sh
> I posted this issue to LibreSSL project, too.
The Cygwin project does not provide or support a libressl package;
there may be good reasons why no package was provided so far:
some upstream projects will not accept a Cygwin patch, so Cygwin
volunteers may decide the package is not long term maintainable,
or adapting the package may need a lot of work, as advised in:
So a fork regression test designed for Unix failed on Windows:
maybe describe what the test does, what fails on Cygwin, and why
that is a problem?
What does "safeness" mean, what does your test log show, and was
it a fork failure?
Does your package build Cygwin dlls, and did you follow recent
advice posted for developers about using rebase on package dlls:
Are you running any https://cygwin.com/acronyms/#BLODA apps shown:
that could interfere with Cygwin fork, mainly intrusive AV/AMW?
Maybe post a https://cygwin.com/acronyms/#STC (Simple Test Case)
using extracts from the test code, the compilation and run output,
and describe the results, so someone can look at it and say if
something needs changed in upstream code, the test can be adjusted
for Cygwin's emulation of Unix fork using Windows calls, if
something in Cygwin can be improved or fixed, or if that's just
the way Cygwin works.
Please read the FAQ and documentation about fork, and follow
the problem reporting guidelines below, and a volunteer Cygwin
maintainer may get time to take a look at the issue, or you
could submit a patch yourself https://cygwin.com/acronyms/#PTC
either upstream or to Cygwin.
Take care. Thanks, Brian Inglis, Calgary, Alberta, Canada
Problem reports: http://cygwin.com/problems.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple