[Openmcl-devel] armcl - fix ucontext_t so we can rebuild-ccl

[R. Matthew Emerson]

The ARM port could benefit from some attention, but it’s not in a terrible state.  There’s a long-standing threading bug that has been extremely difficult to find.  There’s also a bug where it needs abnormally large thread stack sizes in certain circumstances.

The difference between the 1.11.5 and 1.12-dev branches is terribly significant on ARM (whereas there are many more differences on x86).


> On Jul 12, 2018, at 3:38 PM, Ralf Stoye <stoye at stoye.com> wrote:
> 
> Hello everyone,
> 
> to be able to compile the arm kernel i applied the same patch as
> a87d61b88e1f48a563335062668970f7e6290ecf to the platform-linuxarm.h
> Compiling throws some warnings about implicit declared functions. i did
> the same on an own branch based on tag v1.11.5 and got the same warnings.
> 
> Since i want to do some serious things on the raspberry, i would like to
> know:
> - is the arm port stable?
> - should i go with 1.11 or 1.12
> 
> to make things easy, i forked ccl and created branches rs (following
> master) and rs1.115 (following v1.11.5)
> https://github.com/ralfstoye/ccl
> 
> this is just a start to get more involved with the future development of
> ccl,
> Unfortunately i'am not a c-hacker familiar with low level stuff like
> signaling / threading in the context of lisp (gc.....)
> 
> regards
> Ralf
> 
> 
> _______________________________________________
> Openmcl-devel mailing list
> Openmcl-devel at clozure.com
> https://lists.clozure.com/mailman/listinfo/openmcl-devel