Article View: gnu.gcc.bug
Article #1903Re: very wild switch behavior in gcc-1.30
From: Michael Schmidt
Date: Tue, 01 Nov 1988 10:41
Date: Tue, 01 Nov 1988 10:41
19 lines
1559 bytes
1559 bytes
In article <8810271224.AA18236@rikers.ksr.com>, ksr!benson@HARVARD (Benson I. Margulies Kendall Square Research Corp.) writes:
>
>And slower to run. The big cases in here were a deliberate tradeoff of
>maintenance for speed. The Sun compiler handles them just fine with
>the idiot -J argument that demands 32 bit branches. Silly me, I assumed
>that in the absence of such an argument to gcc that either (1) it always
>use long branches, or more likely (2) it used them when the switch
>was big enough to require them. Color me slightly disallusioned.
>
And when you are using gas, you get the following:
michael@hanna <~> gas -v -J
Gnu assembler version 1.18 (I guess.)
gas: I can do better than -J!
--
Michael Schmidt, Universitaet-GH Paderborn, FB 17, Warburger Str.100,
D-4790 Paderborn, West Germany
Mail: michael@pbinfo.UUCP or michael%pbinfo@uunet.uu.net
Technical Details
Message-ID:
<781@pbinfo.UUCP>Path:
arc1-ca.pugleaf.net!utzoo!telly!ddsw1!lll-winken!uunet!mcvax!unido!pbinfo!michaelReferences:
<8810260103.AA06006@ji.Berkeley.EDU> <8810271224.AA18236@rikers.ksr.com>Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!utzoo!telly!ddsw1!lll-winken!uunet!mcvax!unido!pbinfo!michael From: michael@pbinfo.uni-paderborn.de (Michael Schmidt) Newsgroups: gnu.gcc.bug Subject: Re: very wild switch behavior in gcc-1.30 Message-ID: <781@pbinfo.UUCP> Date: 1 Nov 88 10:41:48 GMT References: <8810260103.AA06006@ji.Berkeley.EDU> <8810271224.AA18236@rikers.ksr.com> Sender: news@pbinfo.UUCP Reply-To: michael@pbinfo.uucp Distribution: gnu Organization: Uni-GH Paderborn, FB17/Informatik Lines: 18 In-reply-to: ksr!benson@HARVARD.HARVARD.EDU (Benson I. Margulies Kendall Square Research Corp.)