From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 23440 invoked by alias); 15 Aug 2012 06:40:50 -0000 Received: (qmail 23416 invoked by uid 22791); 15 Aug 2012 06:40:47 -0000 X-SWARE-Spam-Status: No, hits=-1.6 required=5.0 tests=AWL,BAYES_00,RCVD_IN_DNSWL_NONE X-Spam-Check-By: sourceware.org Received: from mo-p00-ob.rzone.de (HELO mo-p00-ob.rzone.de) (81.169.146.161) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Wed, 15 Aug 2012 06:40:33 +0000 X-RZG-AUTH: :LXoWVUeid/7A29J/hMvvT2k715jHQaJercGOZE+TiTS5oCO6h4xP8sw= X-RZG-CLASS-ID: mo00 Received: from [192.168.2.100] (dslb-084-058-207-201.pools.arcor-ip.net [84.58.207.201]) by smtp.strato.de (josoe mo40) (RZmta 30.9 DYNA|AUTH) with ESMTPA id L02553o7F2lB0W ; Wed, 15 Aug 2012 08:40:26 +0200 (CEST) Message-ID: <502B43D1.5040506@gjlay.de> Date: Wed, 15 Aug 2012 07:57:00 -0000 From: Georg-Johann Lay User-Agent: Thunderbird 2.0.0.24 (Windows/20100228) MIME-Version: 1.0 To: binutils@sourceware.org CC: Eric Weddington , Denis Chertykov Subject: [RFC,Patch,AVR]: ad PR13503: Add new relocs to TC_VALIDATE_FIX? Content-Type: multipart/mixed; boundary="------------060305020107050505020807" X-IsSubscribed: yes Mailing-List: contact binutils-help@sourceware.org; run by ezmlm Precedence: bulk List-Id: List-Subscribe: List-Archive: List-Post: List-Help: , Sender: binutils-owner@sourceware.org X-SW-Source: 2012-08/txt/msg00271.txt.bz2 This is a multi-part message in MIME format. --------------060305020107050505020807 Content-Type: text/plain; charset=ISO-8859-15; format=flowed Content-Transfer-Encoding: 7bit Content-length: 1163 Hi, in gas/config/tc-avr.h there is the following comment to TC_VALIDATE_FIX: /* We don't want gas to fixup the following program memory related relocations. We will need them in case that we want to do linker relaxation. We could in principle keep these fixups in gas when not relaxing. However, there is no serious performance penilty when making the linker make the fixup work. Check also that fx_addsy is not NULL, in order to make sure that the fixup refers to some sort of lable. */ PR13503 introduced new relocs that are not mentioned in TC_VALIDATE_FIX. The attached patch adds the new BFD_RELOC_AVR_8_LO, BFD_RELOC_AVR_8_HI, BFD_RELOC_AVR_8_HLO to that exclude list. I cannot say if these new relocs need to be handled there, if the must be handled there or must *not* be handled there. I just noticed that the new relocs are missing there, but no clue if that's good or bad or simply does not matter... Thus please review this patch. And in the case it is actually needed, please apply it. Johann gas/ PR 13503 * config/tc-avr.h (TC_VALIDATE_FIX): Skip: BFD_RELOC_AVR_8_LO, BFD_RELOC_AVR_8_HI, BFD_RELOC_AVR_8_HLO. --------------060305020107050505020807 Content-Type: text/plain; name="bin-pr13503-valid-fix.diff" Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="bin-pr13503-valid-fix.diff" Content-length: 943 diff --git a/gas/config/tc-avr.h b/gas/config/tc-avr.h index 777a9df..e82bd01 100644 --- a/gas/config/tc-avr.h +++ b/gas/config/tc-avr.h @@ -139,11 +139,14 @@ extern long md_pcrel_from_section (struct fix *, segT); || FIXP->fx_r_type == BFD_RELOC_AVR_LO8_LDI_PM_NEG \ || FIXP->fx_r_type == BFD_RELOC_AVR_HI8_LDI_PM_NEG \ || FIXP->fx_r_type == BFD_RELOC_AVR_HH8_LDI_PM_NEG \ + || FIXP->fx_r_type == BFD_RELOC_AVR_8_LO \ + || FIXP->fx_r_type == BFD_RELOC_AVR_8_HI \ + || FIXP->fx_r_type == BFD_RELOC_AVR_8_HLO \ || FIXP->fx_r_type == BFD_RELOC_AVR_16_PM) \ && (FIXP->fx_addsy)) \ { \ goto SKIP; \ - } + } /* This target is buggy, and sets fix size too large. */ #define TC_FX_SIZE_SLACK(FIX) 2 --------------060305020107050505020807--