machine mode for floating point numbers

Article View: gnu.gcc.help

Article #52873

#52873

From: dong
Date: Sat, 25 Apr 1992 10:09

19 lines
957 bytes

My tm.h has the following configurations :

#define BITS_PER_UNIT 32
#define BITS_PER_WORD 32
#define UNITS_PER_WORD 1

#define INT_TYPE_SIZE 32
#define DOUBLE_TYPE_SIZE 64

Can you tell me what is the appropriate mode for a double under
such definition ?

I found gcc used BLKmode for double constants,  so it got
aborted in emit_block_move, as it's not a MEM.

Thanks

Dong

Technical Details

Message-ID: <9204260009.AA13698@vlsi.cs.caltech.edu>

Path: arc1-ca.pugleaf.net!sparky!uunet!zaphod.mps.ohio-state.edu!cis.ohio-state.edu!vlsi.cs.caltech.edu!dong

Show Full Headers

Complete Article Headers

X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 11318c,ee3ba801bad834a1,start
X-Google-Attributes: gid11318c,public
X-Google-ArrivalTime: 1992-04-25 03:09:42 PST
Newsgroups: gnu.gcc.help
Path: arc1-ca.pugleaf.net!sparky!uunet!zaphod.mps.ohio-state.edu!cis.ohio-state.edu!vlsi.cs.caltech.edu!dong
From: dong@vlsi.cs.caltech.edu
Subject: machine mode for floating point numbers
Message-ID: <9204260009.AA13698@vlsi.cs.caltech.edu>
Sender: daemon@cis.ohio-state.edu
Organization: GNUs Not Usenet
Distribution: gnu
Date: Sat, 25 Apr 1992 10:09:42 GMT
Lines: 19