Re: [Chicken-hackers] [PATCH] fix for #1231

[Peter Bex]

On Wed, Feb 10, 2016 at 03:23:23PM +0100, address@hidden wrote:
> Hello!
> 
> 
> Attached is a new version of Jörg's patch for srfi-18, that fixes some 
> problems with
> mutexes. It contains an additional change supplied by Jörg that sets the 
> "blocking object"
> slot in the thread structure correctly. A patch for the srfi-18 egg for 
> chicken5 is attached
> as well. I ran all tests, but (as Peter has already pointed out), our 
> threading tests are
> not particular exhaustive.

Thanks for putting in the effort.  I've taken another look at the patch,
and though I don't fully grok it, the new tests boost confidence that at
least the changes are good.  I've also tagged srfi-18 version 1.1.

During one of my test runs (before applying the patch!) I got the dreaded
exception "Error: (equal?) recursion too deep or circular data encountered"
but it went away the second time I ran it, and also didn't reappear after
patching and rebuilding.

So I think this issue (#1236) deserves some looking into soon :(

I'm not sure we should let this delay the 4.11 release, though I do think
it can be due to the argvector changes and how the stack layout changed
that this is popping up more and more often (it never used to be a
problem, to my knowledge).

Cheers,
Peter

signature.asc
Description: Digital signature