Re: distcc for ada (part 1 of 2)

Thanks for the comments.

I'll re-baseline to the current SVN; might take a week or so since this is a
spare time job for me.

Since the -x option only explicitly tells gcc what the source is supposed to
be, shouldn't distcc already support -x c, -x c++ and -x objective-c? The
reason I added code to support -x is the ada multi-language build tool
"gprbuild" by default adds -x for all languages it supports. Haven't tried
mixing the -x option with the wrong source type.

Agree that max_discrepancies_before_demotion = 10000000 (or whatever large
number) is wrong for c/c++ and ada too, but 1 seems too strict. I think 10
sounds more reasonable, allowing some remote compile failures without
reverting to non-pump compiles. It would be nice to make this user
configurable without recompiling.

Agree about the test comments; I didn't spend enough time trying to
understand the current test approach, and how to integrate with it. I'll
take another look.

Need something to distinguish protocol version 4, but personally, I think a
server version query would be preferable to another suffix on the hostname.
I didn't update the server version query response for version 4; this is a
bug.

I'll write something up in doc/protocol-4.txt for the protocol version 4
changes.

Thanks again,
Tim

__
distcc mailing list            http://distcc.samba.org/
To unsubscribe or change options:
https://lists.samba.org/mailman/listinfo/distcc