This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC 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]

RE: Suspected bug lshift_value (expmed.c) and its patch

BITS_PER_WORD is 64-bit. Shouldn't the description of need_64bit_hwint 
make clear it has to be true if BITS_PER_WORD is 64-bit? If I understand
all others' mails correctly.

Bingfeng

> -----Original Message-----
> From: Paolo Bonzini [mailto:paolo.bonzini@gmail.com] On 
> Behalf Of Paolo Bonzini
> Sent: 19 March 2009 17:48
> To: Bingfeng Mei
> Cc: Andrew Pinski; gcc-patches@gcc.gnu.org
> Subject: Re: Suspected bug lshift_value (expmed.c) and its patch
> 
> 
> >> Is your target a LLP64L32?  I think it should mention 
> pointers there
> > 
> > No.  We have long long (64-bit)  long (32-bit), pointer 
> (32-bit). I checked
> > this description before, but didn't figure out that my 
> target has to use 
> > 64-bit HOST_BITS_PER_WIDE_INT ( a bit lazy to do all 
> cleanup really :-)
> 
> BITS_PER_WORD?  That's the one that counts.
> 
> Paolo
> 
>
Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]