From mboxrd@z Thu Jan 1 00:00:00 1970 X-Msuck: nntp://news.gmane.org/gmane.linux.lib.musl.general/11279 Path: news.gmane.org!.POSTED!not-for-mail From: Szabolcs Nagy Newsgroups: gmane.linux.lib.musl.general Subject: Re: powerpc toolchain question Date: Mon, 24 Apr 2017 11:02:18 +0200 Message-ID: <20170424090218.GT2082@port70.net> References: <1f55bf91-e1d8-8609-d848-8d9141629722@landley.net> Reply-To: musl@lists.openwall.com NNTP-Posting-Host: blaine.gmane.org Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Trace: blaine.gmane.org 1493024552 1867 195.159.176.226 (24 Apr 2017 09:02:32 GMT) X-Complaints-To: usenet@blaine.gmane.org NNTP-Posting-Date: Mon, 24 Apr 2017 09:02:32 +0000 (UTC) User-Agent: Mutt/1.6.0 (2016-04-01) To: musl@lists.openwall.com Original-X-From: musl-return-11294-gllmg-musl=m.gmane.org@lists.openwall.com Mon Apr 24 11:02:27 2017 Return-path: Envelope-to: gllmg-musl@m.gmane.org Original-Received: from mother.openwall.net ([195.42.179.200]) by blaine.gmane.org with smtp (Exim 4.84_2) (envelope-from ) id 1d2Zt5-0000LK-8m for gllmg-musl@m.gmane.org; Mon, 24 Apr 2017 11:02:27 +0200 Original-Received: (qmail 28052 invoked by uid 550); 24 Apr 2017 09:02:31 -0000 Mailing-List: contact musl-help@lists.openwall.com; run by ezmlm Precedence: bulk List-Post: List-Help: List-Unsubscribe: List-Subscribe: List-ID: Original-Received: (qmail 28025 invoked from network); 24 Apr 2017 09:02:30 -0000 Mail-Followup-To: musl@lists.openwall.com Content-Disposition: inline In-Reply-To: <1f55bf91-e1d8-8609-d848-8d9141629722@landley.net> Xref: news.gmane.org gmane.linux.lib.musl.general:11279 Archived-At: * Rob Landley [2017-04-23 19:28:14 -0500]: > I checked that mcm-buildall.sh script (the one I've sent here a few > times) into https://github.com/landley/mkroot and it's building a simple > kernel and root filesystem with the musl-cross-make toolchain that boots > under qemu. So far I've done x86-64, sh4, and powerpc32 kernels. > > But back in Aboriginal Linux, the 32 bit ppc target had gcc building with: > > GCC_FLAGS=--enable-secureplt # --with-long-double-64 (default on 4.2.1) > > I.E. my old 4.2.1 toolchain was doing --enable-secureplt and I had a > comment that newer ones needed --with-long-double-64. I'm not currently > telling the gcc configure to do either of these. Should I be? (Seems to > work but I'm not sure anything's using floating point yet...) --with-long-double-64 should be --without-long-double-128, but it is the default if the gcc configure script cannot detect a recent enough glibc version so it works out. and --enable-secureplt is the default since gcc-6 on powerpc*-linux*-musl* on older gcc you still need it.