RE: [openrisc] eCos & gcc bug

["Robert Cragie" <rcc at jennic dot com>]

>
> > There is an issue in that I have changed a header file. I did a
> look to see
> > if anything called cyg_libc_time_itoa() and AFAICT nothing did.
>
> Actually look at
> ecos/packages/language/c/libc/time/current/include/time.inl to see where
> it's used, although it should be benign, albeit possibly with a warning.

Whoops, sorry about that - forgot to grep .inl files (red face).

> Not surprisingly I'm a little reluctant to apply a patch in generic code
> just because of a toolchain bug. Is it likely this toolchain bug would be
> fixed soon? Has someone checked if it's fixed in more recent GCC?

or32 is not yet in the main gcc trunk - I am using an 'experimental' version
published on the OpenCores site. No-one from OpenCores has responded yet. If
I knew more about GCC porting, I'd tackle it myself but it looks a bit
daunting.

> If it really won't be for a while, then I'll apply the patch since it is
> pretty benign. Once you send a new patch including time.inl and a
> ChangeLog entry anyway ;-).

Patchfile attached.

Robert Cragie, Design Engineer
_______________________________________________________________
Jennic Ltd, Furnival Street, Sheffield, S1 4QT,  UK
http://www.jennic.com  Tel: +44 (0) 114 281 2655
_______________________________________________________________

Attachment: language_c_libc_time.patch
Description: Binary data