Need someone to look at a regression

[Kaveh R. Ghazi]

 > From: Jeffrey A Law <law@hurl.cygnus.com>
 > 
 >   In message < 199808310238.WAA06991@caip.rutgers.edu >you write:
 >   >     Anyway in trying to piece together what's going on, I looked in
 >   > libgcc2.c and noticed that it avoids the warning by doing:
 >   > 
 >   >  > typedef float SFtype __attribute__ ((mode (SF)));
 >   >  > typedef float DFtype __attribute__ ((mode (DF)));
 >   > 
 >   >     So at least to be consistent, I'd like to put 'typedef float'
 >   > into fp-bit.c instead of 'typedef int'.
 > Fine.

	Okay, I'll do that when I can bootstrap the platform.


 >  
 >   >     Sadly, I am unable to test either of the explicit types now
 >   > because in trying to bootstrap the trunk (checked out earlier today) on
 >   > the platform where this all happened (powerpc-ibm-aix4.1.4.0) I'm
 >   > getting major bootstrap problems. 
 > 
 > Presumably this is not during the stage1 compile?  Which would indicate
 > that the compiler has mis-compiled itself at some point.
 > jeff

	Yup, it happened during stage2.

		--Kaveh
--
Kaveh R. Ghazi			Engagement Manager / Project Services
ghazi@caip.rutgers.edu		Icon CMT Corp.