This is the mail archive of the gcc@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: Attacking quadratic behaviors associated with SWITCH_EXPR

From: Kazu Hirata <kazu at cs dot umass dot edu>
To: rakdver at atrey dot karlin dot mff dot cuni dot cz
Cc: gcc at gcc dot gnu dot org, stevenb at suse dot de
Date: Mon, 25 Oct 2004 09:55:03 -0400 (EDT)
Subject: Re: Attacking quadratic behaviors associated with SWITCH_EXPR
References: <20041022.204311.102581980.kazu@cs.umass.edu><20041024080317.GA27324@atrey.karlin.mff.cuni.cz>

Hi Zdenek,

> statement annotations are huge, so this would indeed be a bad idea.
> I think it would be better to avoid using tree for CASE_LABEL_EXPR
> inside switch statement completely.  I.e. TREE_OPERAND (switch, 2)
> would no longer be a TREE_VEC of CASE_LABEL_EXPRs, but instead
> a wrapper tree node (of type tcc_exceptional) containing
> a VEC(case_label_type), where case_label_type would contain just the
> fields necessary for the case label.  This should also save some
> memory over the current approach, since it is a bit less bloated
> representation.

OK, I'll try this route then.

Kazu Hirata

References:
- Attacking quadratic behaviors associated with SWITCH_EXPR
  - From: Kazu Hirata
- Re: Attacking quadratic behaviors associated with SWITCH_EXPR
  - From: Zdenek Dvorak

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]