From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 60712 invoked by alias); 17 Mar 2018 23:01:25 -0000 Mailing-List: contact libc-alpha-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: libc-alpha-owner@sourceware.org Received: (qmail 60702 invoked by uid 89); 17 Mar 2018 23:01:24 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-1.9 required=5.0 tests=BAYES_00,RCVD_IN_DNSWL_NONE autolearn=ham version=3.3.2 spammy= X-HELO: homiemail-a50.g.dreamhost.com Subject: Re: [PATCH] Use correct includes in benchtests To: Florian Weimer , Wilco Dijkstra , "libc-alpha@sourceware.org" Cc: nd References: <4ede91d4-1265-1d43-a10f-6b101969f4f4@redhat.com> From: Siddhesh Poyarekar Message-ID: Date: Sat, 17 Mar 2018 23:01:00 -0000 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.6.0 MIME-Version: 1.0 In-Reply-To: <4ede91d4-1265-1d43-a10f-6b101969f4f4@redhat.com> Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-SW-Source: 2018-03/txt/msg00426.txt.bz2 On Friday 16 March 2018 11:42 PM, Florian Weimer wrote: > On 03/16/2018 07:06 PM, Wilco Dijkstra wrote: >> Florian Weimer wrote: >> >>> I think this broken --enable-static-pie builds for some reason: >>> >>> In file included from bench-timing-type.c:19:0: >>> bench-timing.h:19:0: error: "attribute_hidden" redefined [-Werror] >>> =C2=A0=C2=A0 #define attribute_hidden >>> In file included from :0:0: >>> ./../include/libc-symbols.h:361:0: note: this is the location of the >>> previous definition >>> =C2=A0=C2=A0 # define attribute_hidden __attribute__ ((visibility ("hid= den"))) >>> >>> Any suggestions how to fix this?=C2=A0 Just stick an #undef attribute_h= idden >>> in front of it? There's a bench-timing.h change in f1c8185d345 that wasn't part of the posted patch, that's what is causing the breakage. >> Hmm that looks like a preincluded file, which is done just before we >> define _ISOMAC... So I wonder whether it would be better to fix >> Makeconfig to pre-includes after the CPPFLAGS rather than in the >> middle of it? >=20 > I think Zack's suggestion regarding the change of module will help here. I suppose it would (once someone tries to figure out the nonlib behaviour again) but the easier fix for now should be to just undef it before defining. Siddhesh