From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 31822 invoked by alias); 28 Sep 2009 19:01:15 -0000 Received: (qmail 31812 invoked by uid 22791); 28 Sep 2009 19:01:14 -0000 X-SWARE-Spam-Status: No, hits=-2.4 required=5.0 tests=AWL,BAYES_00,SPF_HELO_PASS,SPF_PASS X-Spam-Check-By: sourceware.org Received: from mx1.redhat.com (HELO mx1.redhat.com) (209.132.183.28) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Mon, 28 Sep 2009 19:01:10 +0000 Received: from int-mx03.intmail.prod.int.phx2.redhat.com (int-mx03.intmail.prod.int.phx2.redhat.com [10.5.11.16]) by mx1.redhat.com (8.13.8/8.13.8) with ESMTP id n8SJ18JM021766; Mon, 28 Sep 2009 15:01:08 -0400 Received: from ns3.rdu.redhat.com (ns3.rdu.redhat.com [10.11.255.199]) by int-mx03.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with ESMTP id n8SJ17VT027990; Mon, 28 Sep 2009 15:01:07 -0400 Received: from opsy.redhat.com (ovpn01.gateway.prod.ext.phx2.redhat.com [10.5.9.1]) by ns3.rdu.redhat.com (8.13.8/8.13.8) with ESMTP id n8SJ16BF011124; Mon, 28 Sep 2009 15:01:06 -0400 Received: by opsy.redhat.com (Postfix, from userid 500) id B7E6D37818C; Mon, 28 Sep 2009 13:01:05 -0600 (MDT) From: Tom Tromey To: Bryce McKinlay Cc: Andrew Haley , Herman ten Brugge , java@gcc.gnu.org Subject: Re: very slow jni code callbacks References: <4ABF6753.20507@home.nl> <4AC0A614.2040108@redhat.com> <7230133d0909280510w1728ce99i1cda1b41fb8f17e5@mail.gmail.com> Reply-To: Tom Tromey Date: Mon, 28 Sep 2009 19:01:00 -0000 In-Reply-To: <7230133d0909280510w1728ce99i1cda1b41fb8f17e5@mail.gmail.com> (Bryce McKinlay's message of "Mon, 28 Sep 2009 13:10:17 +0100") Message-ID: User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/23.1 (gnu/linux) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-IsSubscribed: yes Mailing-List: contact java-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: java-owner@gcc.gnu.org X-SW-Source: 2009-09/txt/msg00035.txt.bz2 Bryce> I think the underlying issue is that jMethodID is typedef'ed to Bryce> libgcj's internal _Jv_Method. It really needs to use something more Bryce> like the interpreter's _Jv_ResolvedMethod or _Jv_JNIMethod so that the Bryce> arg types and return type (and, in fact, the ffi_cif itself) can be Bryce> resolved once and reused. I think there is some reason we use _Jv_Method, and not something else, for the jMethodID typedef. I don't remember what it was. Tom