This is the mail archive of the mailing list for the eCos project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug 1000280] New port: Freescale MAC7100 Variant, Freescale ESCI serial driver

------- Additional Comments From  2006-05-29 20:19 -------
> For that reason I have decoupled serial driver from hal. Serial port macros stay
> together with driver and the statement:
>    requires { CYGPKG_IO_SERIAL_DEVICES == 1 }
>is in order to instruct configtool to include ser_esci.h even in case serial
>port is not selected (by the user).

I'm pritty sure that the header file will be placed into install/include without
needing CYGPKG_IO_SERIAL_DEVICES == 1. There are some cases when you don't want
the serial drivers, eg when building Redboot. If you can make it build without
this that would be great.

>That's simply MAC I/O configuration feature. Every pin has it's own register.
>I was thinking of putting port init. in a loop, but it would require another
>At the time I was beginning the port I knew very little about eCos booting
>process (as well as about eCos itself - actually LED macros were my 1st ever
>eCos port code) and since it states,  in vectors.S, that LED macro can use only
>r0 and r1 I decided to play safe and make stupidest code. I you wish I can loop
>it, but I think that space gain isn't worth it.

Ah, so does strh r1,[r0],#2 do an increment? When i first read the code it just
looked like you were doing nops. So a comment would help explain what is going
on for somebody who does not know the assembly language too well.

Calculated { "ROM" } is O.K. Please change the description to fit. 

Configure bugmail:
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]