From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 13503 invoked by alias); 3 Jan 2018 13:42:43 -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 13292 invoked by uid 89); 3 Jan 2018 13:42:43 -0000 Authentication-Results: sourceware.org; auth=none X-Virus-Found: No X-Spam-SWARE-Status: No, score=-1.5 required=5.0 tests=AWL,BAYES_00,KAM_NUMSUBJECT,RCVD_IN_DNSWL_NONE,SPF_PASS,URIBL_RED autolearn=no version=3.3.2 spammy= X-HELO: relay1.mentorg.com Date: Wed, 03 Jan 2018 13:42:00 -0000 From: Joseph Myers To: Christoph Hellwig CC: Palmer Dabbelt , , Andrew Waterman , Darius Rad , Subject: Re: RISC-V glibc port v2 In-Reply-To: <20180103133714.GA14077@lst.de> Message-ID: References: <20180103133714.GA14077@lst.de> User-Agent: Alpine 2.20 (DEB 67 2015-01-07) MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" X-ClientProxiedBy: svr-ies-mbx-01.mgc.mentorg.com (139.181.222.1) To svr-ies-mbx-01.mgc.mentorg.com (139.181.222.1) X-SW-Source: 2018-01/txt/msg00098.txt.bz2 On Wed, 3 Jan 2018, Christoph Hellwig wrote: > On Mon, Jan 01, 2018 at 01:19:48AM +0000, Joseph Myers wrote: > > Do I understand correctly from this that RV64I processors cannot execute > > RV32I code (unlike e.g. x86_64 where execution of 32-bit code is always > > supported by the processor, or AArch64 where processors may or may not > > support execution of AArch32 code)? If so, then indeed > > bits/environments.h is not relevant at present. > > RV64I processors might be able to execute RV32I code (that support > is optional in the architecture by writing to the sstatus register). But > the Linux port currently does not support that (yet). If suppported in future, then all the multi-ABI pieces I mentioned (ldconfig / ldd support etc.) will become relevant for pieces built with glibc for one of RV64I and RV32I working properly in the presence on the system of libraries for the other one of those two, as well as bits/environments.h. For now, in the absence of such support, those pieces are only relevant for differences in floating-point ABI. -- Joseph S. Myers joseph@codesourcery.com