From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 7190 invoked by alias); 4 Mar 2008 00:10:14 -0000 Received: (qmail 7045 invoked by uid 48); 4 Mar 2008 00:09:30 -0000 Date: Tue, 04 Mar 2008 00:10:00 -0000 Message-ID: <20080304000930.7044.qmail@sourceware.org> X-Bugzilla-Reason: CC References: Subject: [Bug c++/35262] [4.4 Regression]: FAIL: abi_check In-Reply-To: Reply-To: gcc-bugzilla@gcc.gnu.org To: gcc-bugs@gcc.gnu.org From: "pcarlini at suse dot de" Mailing-List: contact gcc-bugs-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-bugs-owner@gcc.gnu.org X-SW-Source: 2008-03/txt/msg00197.txt.bz2 ------- Comment #15 from pcarlini at suse dot de 2008-03-04 00:09 ------- (In reply to comment #14) > Hi, > this is what I get from our thester: > > Differences: > Tests that now work, but didn't before: > abi_check > > so it ought to make differnece for i686-linux. Note however, that the patch also didn't help Geoff's i686-linux tester, just have a look to gcc-testresults. > It is quite possible that things differ on 64bit hosts, we are staying > on quite fragile ground here because in the current cost metric the > benefits of inlining are very close to costs. Given the nature that > function call of the wrapped function is a bit chepaer than call of the > wrapper is quite correct. > > The decision on whether function should be inlined or not depends on > many things, like overall size, ABI details etc. I see it is quite > irritating for ABI checking. I think we should not mix the two issues, here. The first issue is that, IMO, the function we are discussing should be inlined, it's very small and we always inlined it until recently. The other issue is the ABI check failure. I said issue, but actually it's really trivial, just matter of tweaking a bit the linker script, making it a little more tight. I don't think much more is necessary. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=35262