From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 27711 invoked by alias); 2 Nov 2010 22:57:25 -0000 Received: (qmail 27698 invoked by uid 22791); 2 Nov 2010 22:57:22 -0000 X-SWARE-Spam-Status: No, hits=-1.9 required=5.0 tests=AWL,BAYES_00,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,FREEMAIL_FROM,RCVD_IN_DNSWL_NONE,TW_BJ,TW_DL X-Spam-Check-By: sourceware.org Received: from mail-gx0-f169.google.com (HELO mail-gx0-f169.google.com) (209.85.161.169) by sourceware.org (qpsmtpd/0.43rc1) with ESMTP; Tue, 02 Nov 2010 22:57:17 +0000 Received: by gxk2 with SMTP id 2so3653405gxk.0 for ; Tue, 02 Nov 2010 15:57:15 -0700 (PDT) Received: by 10.151.110.2 with SMTP id n2mr1325886ybm.164.1288738635696; Tue, 02 Nov 2010 15:57:15 -0700 (PDT) Received: from bubble.grove.modra.org ([115.187.252.19]) by mx.google.com with ESMTPS id j64sm6610561yha.24.2010.11.02.15.57.11 (version=TLSv1/SSLv3 cipher=RC4-MD5); Tue, 02 Nov 2010 15:57:14 -0700 (PDT) Received: by bubble.grove.modra.org (Postfix, from userid 1000) id 14160170C1EB; Wed, 3 Nov 2010 09:27:07 +1030 (CST) Date: Tue, 02 Nov 2010 22:57:00 -0000 From: Alan Modra To: "H.J. Lu" Cc: Nick Clifton , Cary Coutant , Binutils Subject: Re: Fwd: PATCH: Extend .zdebug section support to binutils and ld Message-ID: <20101102225707.GL26513@bubble.grove.modra.org> Mail-Followup-To: "H.J. Lu" , Nick Clifton , Cary Coutant , Binutils References: <20100722195924.GA4912@caradoc.them.org> <4CC98DA4.7060804@redhat.com> <20101101220923.GD26513@bubble.grove.modra.org> <20101102215926.GK26513@bubble.grove.modra.org> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: User-Agent: Mutt/1.5.20 (2009-06-14) 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: 2010-11/txt/msg00044.txt.bz2 On Tue, Nov 02, 2010 at 03:16:27PM -0700, H.J. Lu wrote: > On Tue, Nov 2, 2010 at 2:59 PM, Alan Modra wrote: > > On Tue, Nov 02, 2010 at 05:59:35AM -0700, H.J. Lu wrote: > >> I don't think they are all caused by my compress patch. > > > > Perhaps not.  I didn't check them all. > > > >> Most of compress failures are due to linker script. I will > >> submit a patch to update linker scripts. > > > > A large number remain due to BFD_COMPRESS/BFD_DECOMPRESS.  objdump -x > > displays bfd->flags.  I think that to fix this properly you should > > Which tests fails because of BFD_COMPRESS/BFD_DECOMPRESS? This one, for example: powerpc-linux +FAIL: PowerPC Test 1, 32 bit elf The list below remain after your linker script update. arm-linux +FAIL: Simple non-PIC shared library arm-linux +FAIL: Simple PIC shared library arm-linux +FAIL: Simple dynamic application arm-linux +FAIL: Simple static application arm-linux +FAIL: Non-pcrel function reference arm-linux +FAIL: Thumb shared library with ARM entry points arm-linux +FAIL: Mixed ARM/Thumb shared library arm-linux +FAIL: Mixed ARM/Thumb dynamic application arm-linux +FAIL: Mixed ARM/Thumb arch5 dynamic application arm-linux +FAIL: TLS shared library arm-linux +FAIL: TLS dynamic application arm-netbsdelf +FAIL: Simple non-PIC shared library arm-netbsdelf +FAIL: Simple PIC shared library arm-netbsdelf +FAIL: Simple dynamic application arm-netbsdelf +FAIL: Simple static application arm-netbsdelf +FAIL: Non-pcrel function reference arm-netbsdelf +FAIL: Thumb shared library with ARM entry points arm-netbsdelf +FAIL: Mixed ARM/Thumb shared library arm-netbsdelf +FAIL: Mixed ARM/Thumb dynamic application arm-netbsdelf +FAIL: Mixed ARM/Thumb arch5 dynamic application arm-netbsdelf +FAIL: TLS shared library arm-netbsdelf +FAIL: TLS dynamic application arm-none-eabi +FAIL: Simple non-PIC shared library arm-none-eabi +FAIL: Simple PIC shared library arm-none-eabi +FAIL: Simple dynamic application arm-none-eabi +FAIL: Simple static application arm-none-eabi +FAIL: Non-pcrel function reference arm-none-eabi +FAIL: Thumb shared library with ARM entry points arm-none-eabi +FAIL: Mixed ARM/Thumb shared library arm-none-eabi +FAIL: Mixed ARM/Thumb dynamic application arm-none-eabi +FAIL: Mixed ARM/Thumb arch5 dynamic application arm-none-eabi +FAIL: TLS shared library arm-none-eabi +FAIL: TLS dynamic application arm-none-eabi +FAIL: Mixed ARM/Thumb dynamic application with farcalls arm-none-eabi +FAIL: Mixed ARM/Thumb arch5 dynamic application with farcalls dlx-elf +FAIL: ld-elf/compress1a dlx-elf +FAIL: ld-elf/compress1b hppa64-hp-hpux11.11 +FAIL: ld-elf/compress1a hppa64-hp-hpux11.11 +FAIL: ld-elf/compress1b hppa64-hp-hpux11.23 +FAIL: ld-elf/compress1a hppa64-hp-hpux11.23 +FAIL: ld-elf/compress1b i370-linux +FAIL: ld-elf/compress1c microblaze-elf +FAIL: ld-elf/compress1a microblaze-elf +FAIL: ld-elf/compress1b mips64-linux +FAIL: ELF e_flags: nothing special mips64-linux +FAIL: ELF e_flags: -m4650 mips64-linux +FAIL: ELF e_flags: -march=4650 -mtune=4650 mips64-linux +FAIL: ELF e_flags: -march=4650 -mtune=4650 mips64-linux -FAIL: ld-mips-elf/no-shared-1-n64 mips64-linux -FAIL: PIC and non-PIC test 6 (n64 executable) mips64-linux -FAIL: GOT dump (readelf -A) test 2 mipsel-linux-gnu +FAIL: ELF e_flags: nothing special mipsel-linux-gnu +FAIL: ELF e_flags: -m4650 mipsel-linux-gnu +FAIL: ELF e_flags: -march=4650 -mtune=4650 mipsel-linux-gnu +FAIL: ELF e_flags: -march=4650 -mtune=4650 mipsisa32el-linux +FAIL: ELF e_flags: nothing special mipsisa32el-linux +FAIL: ELF e_flags: -m4650 mipsisa32el-linux +FAIL: ELF e_flags: -march=4650 -mtune=4650 mipsisa32el-linux +FAIL: ELF e_flags: -march=4650 -mtune=4650 mips-linux +FAIL: ELF e_flags: nothing special mips-linux +FAIL: ELF e_flags: -m4650 mips-linux +FAIL: ELF e_flags: -march=4650 -mtune=4650 mips-linux +FAIL: ELF e_flags: -march=4650 -mtune=4650 mmix +FAIL: ld-mmix/undef-3m ns32k-netbsd +FAIL: align2 pdp11-dec-aout +FAIL: align2 powerpc64-linux +FAIL: PowerPC Test 1, 64 bit elf powerpc64-linux +FAIL: Power4 instructions powerpc-eabisim +FAIL: PowerPC Test 1, 32 bit elf powerpc-linux +FAIL: PowerPC Test 1, 32 bit elf sh64-elf +FAIL: SH64 linking, 64-bit ABI sh64-elf +FAIL: SH64 linking, 32-bit ABI sh64-elf +FAIL: SH64 linking, single multi-ISA object sh64-elf +FAIL: SH64 linking, two different-ISA objects sh64-elf +FAIL: SH64 linking, single SHcompact sh64-elf +FAIL: SH64 inter-file datalabel references, 64-bit ABI sh64-elf +FAIL: SH64 inter-file datalabel references, 32-bit ABI tic30-unknown-aout +FAIL: align2 tx39-elf +FAIL: ELF e_flags: nothing special tx39-elf +FAIL: ELF e_flags: -m4650 tx39-elf +FAIL: ELF e_flags: -march=4650 -mtune=4650 tx39-elf +FAIL: ELF e_flags: -march=4650 -mtune=4650 > > > store the new compress controls in two single-bit fields.  (See end of > > struct bfd.) > > > > It is easier to use the flags field. Otherwise, I need to update > more places. The alternative of fixing all the failing testcases will take time too. Your choice. From a design/maintenance view I don't really care too much. bfd->flags already has some fields that aren't properties of the bfd, BFD_TRADITIONAL_FORMAT and BFD_DETERMINISTIC_OUTPUT, so I guess smuggling your two controls there doesn't matter too much. Also, I'm not too worried about the objdump -f / objdump -x display change apart from its impact on the testsuite. I haven't checked for any impact on the gdb testsuite. -- Alan Modra Australia Development Lab, IBM