Re: [cc65] some patches

From: Kajtár Zsolt <kajtarzsolt1googlemail.com>
Date: 2010-03-15 20:56:11
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

> On Sonntag 14 März 2010, you wrote:
> 
>> ide64-read-write.patch
>> # Add blockread/write support if IDE64 is present.
>> # This makes a big difference in I/O speed.
>> # Only for C64, to not bloat the other archs.
>>
>> This last patch makes a copy of read.s, write.s and rwcommon.s to the c64
>> directory. This is not nice, but I haven't found an easy way to not have
>>  the IDE64 code compiled in for other cbm machines.
> 
> urks. that means every c64 program would get this - normally unneeded - ide64 
> stuff compiled in? rather looks to me as if this should go in a seperate 
> library.

Only if they are using read/write. But if it's a separate library, normal
programs using read/write (fread/fwrite) will not be accelerated without
modification.

Is it a good idea to create some header file, which would redefine
read/write/fread/fwrite when included? (something like #define read(a,b,c)
read_ide64(a,b,c) ) But then I have to make a copy from fread/fwrite as well...

By the way is it possible to get rid of the interrupt handling stuff in crt0
if I'm not using any interrupts in a program? I've tried to put .if there, but
unfortunately __INTERRUPTOR_COUNT__ is imported and not defined, so it was not
working.

- --
						    -soci-
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)

iEYEARECAAYFAkuekNsACgkQyBOVYiJltRYVDwCfToDI32sWh5W0vPJVP770sjq+
AfgAnRBxvobFvBaS0b55wuB7yZ+mkieM
=yP+r
-----END PGP SIGNATURE-----
----------------------------------------------------------------------
To unsubscribe from the list send mail to majordomo@musoftware.de with
the string "unsubscribe cc65" in the body(!) of the mail.
Received on Mon Mar 15 20:58:25 2010

This archive was generated by hypermail 2.1.8 : 2010-03-15 20:58:27 CET