Propeller GCC Status
jazzed
Posts: 11,803
Ken Gracey asked me to provide a periodic update for the Propeller 2 GCC effort.
Let's all remember that Propeller 2 is an unofficial name; I also call it P2. Sometimes I refer to the current Propeller as P1 or P8x32a like I shoud. I get weary and lost saying Propeller and Propeller 2.
Since the P2 GCC project was announced, we have gathered team members for the development effort. We have been discussing Product Requirements and deliverables lightly until all the paperwork and walnuts are in order. Ken Gracey gave us a charter which includes some requirements, but more will be added.
Some Updates
Two things may be possible that were not part of the original requirement. I have discussed these possibilities briefly with Ken.
Calling all GCC testers!
We will be needing testers as the product matures. Propeller GCC is still a twinkle in our eyes at this point, but once it's learned to walk, we will be getting serious about recruiting more early testers so we can exercise this baby once it can run.
The details of all this are to be determined. There will be a limited supply of P2s for testers, so if you're serious and have some background with GCC, speak up!
Cheers to all!
--Steve
Let's all remember that Propeller 2 is an unofficial name; I also call it P2. Sometimes I refer to the current Propeller as P1 or P8x32a like I shoud. I get weary and lost saying Propeller and Propeller 2.
Since the P2 GCC project was announced, we have gathered team members for the development effort. We have been discussing Product Requirements and deliverables lightly until all the paperwork and walnuts are in order. Ken Gracey gave us a charter which includes some requirements, but more will be added.
Some Updates
Two things may be possible that were not part of the original requirement. I have discussed these possibilities briefly with Ken.
- A Native PASM COG code generator will probably be one of the first things created. You will most likely be able to write your PASM drivers in C.
- A P1 tool chain. The P1 P8x32a is essentially a sub-set of the P2. While it is not designed specifically for the LMM code like the P2, it can run LMM successfully as Bean, ImageCraft, Ross and others have shown.
Calling all GCC testers!
We will be needing testers as the product matures. Propeller GCC is still a twinkle in our eyes at this point, but once it's learned to walk, we will be getting serious about recruiting more early testers so we can exercise this baby once it can run.
The details of all this are to be determined. There will be a limited supply of P2s for testers, so if you're serious and have some background with GCC, speak up!
Cheers to all!
--Steve
Comments
-Phil
Most Excellent.
C.W.
Say what! Does this mean I can't refer to them as the PC1 and PC2 without being corrected ;-)
Whack, whack with a scope probe! Actually, I can't berate you too firmly for using such terms when they are also common inside of Parallax. At a recent company meeting Jen addressed the subject of shorter names for Parallax Semiconductor. Inside we tend to use ParaSemi, but never PSemi since it belongs to other companies and products. She was simply engraving our brains before the PSemi name became commonplace.
Jazzed, thanks for making the update. Without your involvement we'd have a very difficult time getting this project underway.
Ken Gracey
Of course, actually building unit tests for the compiler that interfaces to hardware shouldn't be that hard either.
As always, I am available to be "guinea pigs".
Even if my I'm not a pro in prop
JM
- Agreements for consultants are being signed this week.
- Dave Hein's excellent spinsim simulator is being expanded for Propeller 2 use.
- Development hardware access plan is in progress.
- Requirements Document is still in process.
We still need more tester and demo application volunteers.I would like to say thanks to Luis Digital and jmspaggi for their interest in alpha testing.
Answering questions about Java: the GNU Compiler Collection includes GCJ. It is not clear at this point if GCJ will be supported. It and other languages such as Objective C, Fortran, and Ada may be considered. The main emphasis however at this point is C/C++.
Some initial discussion on an Open Source Hardware Propeller 2 Demo board have started in the forum. It is not an officially sanctioned project ... yet. I hope that all the excellent hardware talent we have here can contribute to that thread.
Massimo
I'm not sure I was clear, but I will be very happy to be part of the testers
And as yhou know, I will be volonteer to test or try to adapt the GCJ procedure
Having another language than spin will be a real pleasure (as long as it's not emulated...)
Thanks!
JM
JM, I got ya
Please add me to the list of potential testers.
Thanks,
C.W.
Sign me up for testing/documenting/sweeping up afterwards. I'm not a professional micro-engineer and I'm not a GCC person at all but I've been programming, using tools, reading instructions and trying to make all different sorts of computer things work for a long, long time. Maybe I can contribute somewhere along the line.
Thanks!
I will already be happy with GCC anyway
JM
There has been an official change in the Propeller GCC project.
We are targeting Propeller 1 (today's PX832A) for GCC first.
Then we will port that to Propeller 2.
This makes sense because PX832A hardware is shipping now.
Customers who demand GCC will get it for both chips.
GCC porting work is in progress.
Daniel Harris at Parallax is now on the GCC project.
Thanks!
--Steve
Do you have any ETA for the first beta release?
JM
A very sensible change. I always thought that it was a bit ambitious to start writing a code generator for a chip that doesn't even have an instruction set yet. Also, we know quite a lot about how to do LMM on the Prop I, so you're not just shooting in the dark and hoping things work out when the chip is eventually released.
Ross.
Yes, I always wondered about that particular decision - especially since 90% of the work has already been done for you on the Prop I (sorry, couldn't resist!
Ross.
Just a quick note to say thanks for the regular updates.
I'm really at "user" level for this sort of thing so not much use contributing to the development path, but I know enough to like the direction its heading.
cheers
Lachlan
Bah, someone is trying to make Zog redundant. It won't work I tell you.
On the other hand:
Sounds like a emminantly sensible idea. Not very practically useful on the Prop I but a solid stepping stone to the goal.
I think this is key.
We keep calling the new chip under development "Propeller 2" which makes people think that it's going to replace our original prop. It's kinda interesting we do that since we don't do the same with the PIC number 76 or AVR number 432 :-)
Having both chips use the same toolchain means that I can choose the right hardware for a specific design without having to worry about using different languages or toolchains.
don't worry, Zog will still have a place - for some programs, small code size is more imporant than raw speed
And when you do need raw speed, Catalina will be there
Ross.
I also suspect that Catalina will beat the GCC port on compilation speed as LCC is leaner/meaner then GCC/bintools.