Article View: gnu.gcc.help
Article #49712Where does gcc evaluate compile time expressions?
From: Paul Budnik uunet!mtnmath!paul
Date: Fri, 31 Dec 1993 05:07
Date: Fri, 31 Dec 1993 05:07
13 lines
1212 bytes
1212 bytes
I have ported gcc to a 16 bit DSP processor and am having problems with optimization. A floating point comparison is optimized away but with the wrong result. The rtl conditional expression occurs in the '.rtl' dump but has been optimized away (incorrectly) in the '.jump' dump. I suspect I can track down the problem if I can find where gcc evaluates compile time expressions but am having difficulty figuring out where this is done. Because this machine cannot do byte addressing the modes are nonstandard. For example 32 bit IEEE floating point operands are HF mode. Perhaps this is the related to the problem. Paul Budnik
Technical Details
Message-ID:
<1993Dec31.050747.646@mtnmath.UUCP>Path:
arc1-ca.pugleaf.net!gmd.de!xlink.net!howland.reston.ans.net!cs.utexas.edu!uunet!mtnmath!paulShow Full Headers
Complete Article Headers
X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: 11318c,c68244d35ccf15ea,start X-Google-Attributes: gid11318c,public X-Google-ArrivalTime: 1993-12-30 23:53:08 PST Newsgroups: gnu.gcc.help Path: arc1-ca.pugleaf.net!gmd.de!xlink.net!howland.reston.ans.net!cs.utexas.edu!uunet!mtnmath!paul From: paul@mtnmath.UUCP (Paul Budnik uunet!mtnmath!paul) Subject: Where does gcc evaluate compile time expressions? Organization: Mountain Math Software, P. O. Box 2124, Saratoga. CA 95070 Date: Fri, 31 Dec 1993 05:07:47 GMT Message-ID: <1993Dec31.050747.646@mtnmath.UUCP> Lines: 13