Re: [Sdcc-user] Super 6502s, was Re: Adding support for the 6502 processor

2012-09-05 Thread Dave McGuire
On 09/05/2012 02:29 PM, Groepaz wrote: > On Wednesday 05 September 2012, Dave McGuire wrote: >> On 09/05/2012 06:52 AM, Groepaz wrote: >>> err, no - the PCE uses a NEC "HuC6280" CPU, which is "6502 like", but >>> contains a bunch of rather specific extension which can not be found in >>> any other

Re: [Sdcc-user] Super 6502s, was Re: Adding support for the 6502 processor

2012-09-05 Thread Groepaz
On Wednesday 05 September 2012, Dave McGuire wrote: > On 09/05/2012 06:52 AM, Groepaz wrote: > > err, no - the PCE uses a NEC "HuC6280" CPU, which is "6502 like", but > > contains a bunch of rather specific extension which can not be found in > > any other CPU (eg block copy, paging support) > >

[Sdcc-user] Super 6502s, was Re: Adding support for the 6502 processor

2012-09-05 Thread Dave McGuire
On 09/05/2012 06:52 AM, Groepaz wrote: > err, no - the PCE uses a NEC "HuC6280" CPU, which is "6502 like", but > contains > a bunch of rather specific extension which can not be found in any other CPU > (eg block copy, paging support) A 6502 with block copy and paging?? Wow. Are these chips

Re: [Sdcc-user] Adding support for the 6502 processor

2012-09-05 Thread Philipp Klaus Krause
On 05.09.2012 19:19, Masur Jonathan wrote: > There is also the 65C816 which has a 16-bit mode and many extra > instructions, which is used in the Super Nintendo Entertainment System > (SNES) video game console. > > It would be great to port SDCC for all variants of the 6502 family, > although

Re: [Sdcc-user] Adding support for the 6502 processor

2012-09-05 Thread Masur Jonathan
Le 05.09.2012 12:52, Groepaz a écrit : > > err, no - the PCE uses a NEC "HuC6280" CPU, which is "6502 like", but contains > a bunch of rather specific extension which can not be found in any other CPU > (eg block copy, paging support) > > an actual 65C02 can be found in the NES aka famicom :) > > t

Re: [Sdcc-user] Adding support for the 6502 processor

2012-09-05 Thread Jan Waclawek
During the years I came across chips designated 65C02 which did not have the bit manipulation etc. "new" instructions. Clearly, there were several second-sourcing companies, some of them might have chosen to designate a 6502-like design as "C" simply to stress the CMOS technology. Also, accordi

Re: [Sdcc-user] Adding support for the 6502 processor

2012-09-05 Thread Groepaz
On Wednesday 05 September 2012, Masur Jonathan wrote: > > Also, as a remark, the 6502 has a derivative, the 65C02; of which > > there also exists variants with "secret" (more or less officially > > supported) instructions. > > Exact, the 65C02 has some extra instructions. It is used in the > Tu

Re: [Sdcc-user] Adding support for the 6502 processor

2012-09-05 Thread Masur Jonathan
Le 05.09.2012 09:32, Sebastien Lorquet a écrit : > Hello, > > It depends on the port. > > [...] > > For your port, you can have both (as for mcs51): functions flagged with > __reentrant will use the stack convention, the others will use the overlay > convention. > Yes, this sounds like the way to

[Sdcc-user] SDCC-3.2.0, 8051 port, problem with inline function.

2012-09-05 Thread Krishnendu Chatterjee
Hi folks, I am facing some problem using inline functions on mcs51 port. Consider the following code: #include int n[10]; inline int *address (void) { return &n[0]; } inline int size (void) { return sizeof (n); } void main (void) { memset (address(), 0, size()); } When it is compiled wi

Re: [Sdcc-user] Adding support for the 6502 processor

2012-09-05 Thread Jan Waclawek
Isn't the HC08 be similar enough to 6502? Also, as a remark, the 6502 has a derivative, the 65C02; of which there also exists variants with "secret" (more or less officially supported) instructions. It might be also interesting to find out whether WDC (http://www.westerndesigncenter.com/wdc/ )

Re: [Sdcc-user] Adding support for the 6502 processor

2012-09-05 Thread Sebastien Lorquet
Le 04/09/2012 20:55, Masur Jonathan a écrit : > I am under the impression this is exactly what SDCC does - not using an > argument stack for maximal code optimisation - at the cost to not being > able to write re-entrant functions. But this is a very small price to > pay for something that can