Re: suspicious code in remake.c

[Boris Kolpackov]

Paul D. Smith <address@hidden> writes:

> This rings a bell with me.  Ah yes, I see you sent this back on 10 Feb.

It is really surprising how you guessed the relation even though I never
sent a patch ;-).

This additional if() proved to be important when I have two other
patches installed: one for dontcare flag inheritance and the other
is for implicit phony suppression. I can't seem to find an example
without those patches that will make a difference (but it doesn't
mean there is no such example ;-) ). 

> I don't know that it's really a problem since f->prev will be NULL for
> single-colon targets and the loop will fall through. 

It is only the case when f = f->prev in the previous loop has been 
executed. Look at all those 'break's in that loop that leave 'f'
non-null. 

> But, I'll check it out.

Thanks,
-boris

signature.asc
Description: Digital signature