Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 8 Aug 1998 08:57:58 +1000 (EST)
From:      John Birrell  <jb@cimlogic.com.au>
To:        jasone@canonware.com (Jason Evans)
Cc:        jb@cimlogic.com.au, current@FreeBSD.ORG
Subject:   Re: SMP and kernel threads (Was: Re: Pthreads woes revisited.)
Message-ID:  <199808072257.IAA11353@cimlogic.com.au>
In-Reply-To: <Pine.BSF.3.96.980807150309.2126C-100000@orkan.canonware.com> from Jason Evans at "Aug 7, 98 03:25:52 pm"

next in thread | previous in thread | raw e-mail | index | archive | help
Jason Evans wrote:
> 1) Which message from John Dyson are you referring to?  I don't recall a
> technically detailed message from him on this subject for 2+ months.

The message I am referring to was dated Fri, 7 Aug 1998 16:12:13 -0500 (EST)
and sent as part of the original thread. I can forward you a copy
if necessary.

> 2) Where are John Dyson's patches currently available?

Send email to John.

> 3) I thought that John Dyson's patches had been committed long ago.  Were
> they not committed because of stability problems, performance problems,
> lack of a warm body to do it, etc.?

AFAIK, they were his work in progress. He wasn't ready to commit them
because he hadn't completed what he wanted to achieve.

> You request that I "write regression testing code to prove that it works". 
> Regression testing does nothing more than prevent old bugs from
> reappearing.  It proves nothing other than that old bugs haven't
> reappeared, and that assumes that the regression tests are written
> correctly to test for the bugs rather than their symptoms. 
> 
> Don't get me wrong; I'm a proponent of regression testing, but it won't
> achieve what you're asking for.  Are you just asking for stability testing
> of the patches, or something else in addition? 

OK maybe "regression" is the wrong word. I'm asking for code that proves
the threads work - code that can be used to ensure that the functionality
hasn't been broken and code that can have additional tests added when
people report bugs. This sort of code is needed so that regression tests
can be performed when the bugs are fixed.

One of the problems I have with FreeBSD is that developer testing is
performed privately and test code if it exists at all is not contributed.
In this case, I want to be able to look at the test code too.

-- 
John Birrell - jb@cimlogic.com.au; jb@freebsd.org http://www.cimlogic.com.au/
CIMlogic Pty Ltd, GPO Box 117A, Melbourne Vic 3001, Australia +61 418 353 137

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-current" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199808072257.IAA11353>