ANNOUNCE: Start Ajhc project with forking jhc.
Hi all. I am a user of jhc Haskell compiler. Jhc can compile Haskell code to micro arch such as Cortex-M3. I have written LED blinking demo for Cortex-M3 with jhc. Very fun! https://github.com/ajhc/demo-cortex-m3 http://www.youtube.com/watch?v=3R9sogReVHg And I created many patches for jhc. But...I think that the upstream author of jhc, John Meacham, can't pull the contribution speedy, because he is too busy. It's difficult that maintain many patches without any repositories, for me. Then, I have decided to fork jhc, named Ajhc. # pain full... http://ajhc.github.com/ I will feedback Ajhc's big changes to jhc mailing list. Or I am so happy if John joins Ajhc project. Regards, -- Kiwamu Okabe
Super cool. I'd actually like to continue to merge your changes back
into the original branch if that's okay. Targeting embedded processors
is a major goal of jhc but I did not really have hardware to play
with. Is there no way you can use darcs to continue development? it
makes pulling patches back and forth between branches very easy.
Though, I suppose i can use git diff to get at your changes, it will
get harder if they diverge much more.
John
On Wed, Mar 6, 2013 at 4:44 AM, Kiwamu Okabe
Hi all.
I am a user of jhc Haskell compiler. Jhc can compile Haskell code to micro arch such as Cortex-M3. I have written LED blinking demo for Cortex-M3 with jhc. Very fun!
https://github.com/ajhc/demo-cortex-m3 http://www.youtube.com/watch?v=3R9sogReVHg
And I created many patches for jhc. But...I think that the upstream author of jhc, John Meacham, can't pull the contribution speedy, because he is too busy. It's difficult that maintain many patches without any repositories, for me.
Then, I have decided to fork jhc, named Ajhc. # pain full...
I will feedback Ajhc's big changes to jhc mailing list. Or I am so happy if John joins Ajhc project.
Regards, -- Kiwamu Okabe
_______________________________________________ jhc mailing list jhc@haskell.org http://www.haskell.org/mailman/listinfo/jhc
Hi John.
On Wed, Mar 6, 2013 at 10:09 PM, John Meacham
Super cool. I'd actually like to continue to merge your changes back into the original branch if that's okay.
I am so happy that merge Ajhc's changing to jhc.
with. Is there no way you can use darcs to continue development? it makes pulling patches back and forth between branches very easy. Though, I suppose i can use git diff to get at your changes, it will get harder if they diverge much more.
Umm.....sorry.... I would like to use git below reasons. 1. To use travis-ci. https://travis-ci.org/ajhc/ajhc 2. To collaborate other people to develop Ajhc. Ajhc's main branch is jhc's darcs repo daily mirror. Out changes are putted on Ajhc's arafura branch. Is there method to mirror git arafura branch => darcs repo? And it's good for your developing style? My future plans are below. ("v" is done. "*" is yet.) v Try to rewrite a part of NetBSD bootloader. http://www.youtube.com/watch?v=JMyxy5Qo2m8 v Translate Jhc User’s Manual to Japanese. http://metasepi.masterq.net/posts/2013-01-12-jhc_manual_ja.html * Fix bug that run stm32f3-discovery demo. It causes Ajhc RTS heep impl. https://github.com/ajhc/demo-cortex-m3/tree/master/stm32f3-discovery * Rewrite Cortex-M3 base library with Ajhc. * Pass all regress test, and enable regress fail setting on travis-ci. https://travis-ci.org/ajhc/ajhc * Analyze jhc internal (Japanese doc). http://metasepi.masterq.net/posts/2013-01-31-jhc_internal_overview.html http://metasepi.masterq.net/posts/2013-02-14-jhc_grin_to_c.html * Find critical region of Ajhc for reentrant and thread safe. * Understand jhc's region inference. * Find the method to implement GC that can be interrupted. * Rewrite NetBSD kernel with Ajhc. Best regards, -- Kiwamu Okabe
participants (2)
-
John Meacham -
Kiwamu Okabe