[{"id":1761109,"web_url":"http://patchwork.ozlabs.org/comment/1761109/","msgid":"<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>","list_archive_url":null,"date":"2017-08-31T15:23:41","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":22438,"url":"http://patchwork.ozlabs.org/api/people/22438/","name":"Carlos O'Donell","email":"carlos@redhat.com"},"content":"On 08/31/2017 04:52 AM, Florian Weimer wrote:\n> Since glibc 2.24, __malloc_initialize_hook is a compat symbol. As a\n> result, the link editor does not export a definition of\n> __malloc_initialize_hook from the main program, so that it no longer\n> interposes the variable definition in libc.so. Specifying the symbol\n> version restores the exported symbol.\n> \n> 2017-08-31 Florian Weimer \n> \n> \t[BZ #22050]\n> \t* malloc/mcheck-init.c (__malloc_initialize_hook): Use\n> \tcompat_symbol_reference to access non-default version.\n>\n> diff --git a/malloc/mcheck-init.c b/malloc/mcheck-init.c\n> index 6d2492ef7e..4ebfa868ea 100644\n> --- a/malloc/mcheck-init.c\n> +++ b/malloc/mcheck-init.c\n> @@ -20,6 +20,7 @@\n> \n> #include \n> #include \n> +#include \n> \n> static void\n> turn_on_mcheck (void)\n> @@ -28,3 +29,7 @@ turn_on_mcheck (void)\n> }\n> \n> void (*__malloc_initialize_hook) (void) = turn_on_mcheck;\n> +/* Slight abuse of compat_symbol_reference because the above is not a\n> + reference, but actually a definition. */\n> +compat_symbol_reference (libc, __malloc_initialize_hook,\n> + __malloc_initialize_hook, GLIBC_2_0);\n> \n\nWe deprecated __malloc_initialize_hook in libc.so.6. There is only a compat\nsymbol there, which can only be interposed by a definition of the same version.\n\nIn libmcheck.a we have __malloc_initialize_hook (unversioned) which no longer\ninterposes the symbol in libc.so.6 (it would have interposed the default if\nthere was one).\n\nSo what does the compat_symbol_reference create in the libmcheck.a? An undefined\nreference to __malloc_initialize_hook@GLIBC_2_0?","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-83932-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"AqHro8kR\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xjmPK5S5Tz9s83\n\tfor ;\n\tFri, 1 Sep 2017 01:24:05 +1000 (AEST)","(qmail 52373 invoked by alias); 31 Aug 2017 15:23:57 -0000","(qmail 52320 invoked by uid 89); 31 Aug 2017 15:23:56 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=dBt3WkgsDYNYTpUL\n\t82TSDEdfSATfGzB5aKqMqRICKwh3fJ1HorOn8IuxIEzj0e10kpYJ+4nGUenmjVny\n\tV7hWNgi0ywTu06isLs1qqXZ/EGbfc4gY4BkAVRUyoAVHP+1ImvP0HwRHJkviC4t6\n\tl3BgHIHZRTXS6X/rD/jQnud+tS0=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; s=default; bh=zLCkJw7JnIZvVDCfUsPKxp\n\tWoqik=; b=AqHro8kRHmOeWRWq9RGLYls6SKpM6ZZFnFanl2MTjcnRaihpPoVYVj\n\tf0E8YwHQYzIIDaXZAMlx6uoB8xBjkdawRm/UmdpdwgU5LjJtVuMylIbJP5fieRYJ\n\t43GrtNbpY8qYA4arxYKyBONVnVoJEvLOR1+UqHMk2pCiq49oRCmY0=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-23.9 required=5.0 tests=AWL, BAYES_00,\n\tGIT_PATCH_0, GIT_PATCH_1, GIT_PATCH_2, GIT_PATCH_3,\n\tRCVD_IN_DNSWL_NONE,\n\tRCVD_IN_SORBS_SPAM autolearn=ham version=3.3.2 spammy=","X-HELO":"mail-qk0-f176.google.com","X-Google-DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed;\n\td=1e100.net; s=20161025;\n\th=x-gm-message-state:subject:to:references:from:organization\n\t:message-id:date:user-agent:mime-version:in-reply-to\n\t:content-language:content-transfer-encoding;\n\tbh=T9nqPK8hCe7ZKfXKhEmPTci/stOA3D651AZY6kShvi0=;\n\tb=Xv9ciQdfXWNHzLjdnAi49iQiPhRo0kgjK4DcbSIWsiYf+YjodScmJVuIdAkdtx6mC5\n\tzOAbTrXFKXPBW09Qdw2EEB4YkMNiMwgcX7Zck8u6O1OB9PO9t8nu9sLxVpXHTJt0vwSU\n\tFJx8APY3Icw86rXFLttcsM2mPXEYxKY8fSFrHtKjs8Hmk4KwBoY28A7qIQq4K6ztM0ah\n\t8oh/1wsGazQWzFu+olAFLD/CQ5VXozIERGlBydCSsHpCVi02MDh/Wuf6l+e4zKT9z5H9\n\tAzUP0pnlyMSgSbBzWS5TBvHEyCTJifzItuZDNa6hBXUdfu8U1mevY1m6KOBpPSB8M/oZ\n\twDzw==","X-Gm-Message-State":"AHYfb5jhVNQBEpvBkfryzDp2rSIpG8lXcazH5TcDSxf4W0CJcfHTSPNY\n\t1oBMAzwobxVV22bFJ64LnQ==","X-Google-Smtp-Source":"ADKCNb6WfrqNANkOfpAfzRjtj9+Jk6ApJqelFiw3jgjnFAoF9+MasH3ToHn54ggAA5ImoahtkqeJ6g==","X-Received":"by 10.55.161.214 with SMTP id k205mr3991204qke.188.1504193024662;\n\tThu, 31 Aug 2017 08:23:44 -0700 (PDT)","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","To":"Florian Weimer , libc-alpha@sourceware.org","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>","From":"Carlos O'Donell ","Message-ID":"<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>","Date":"Thu, 31 Aug 2017 10:23:41 -0500","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.2.1","MIME-Version":"1.0","In-Reply-To":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>","Content-Type":"text/plain; charset=utf-8","Content-Transfer-Encoding":"7bit"}},{"id":1761119,"web_url":"http://patchwork.ozlabs.org/comment/1761119/","msgid":"","list_archive_url":null,"date":"2017-08-31T15:28:27","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":14312,"url":"http://patchwork.ozlabs.org/api/people/14312/","name":"Florian Weimer","email":"fweimer@redhat.com"},"content":"On 08/31/2017 05:23 PM, Carlos O'Donell wrote:\n\n> We deprecated __malloc_initialize_hook in libc.so.6. There is only a compat\n> symbol there, which can only be interposed by a definition of the same version.\n> \n> In libmcheck.a we have __malloc_initialize_hook (unversioned) which no longer\n> interposes the symbol in libc.so.6 (it would have interposed the default if\n> there was one).\n> \n> So what does the compat_symbol_reference create in the libmcheck.a? An undefined\n> reference to __malloc_initialize_hook@GLIBC_2_0?\n\nInterposition happens only if __malloc_initialize_hook is listed in the\n.dynsym section of the executable. At least some versions of binutils\nwill not add the symbol to the .dynsym section if it is unversioned and\nthere is a definition in a DSO which lacks a default version. Arguably\nthis is a bug in the link editor.\n\nThe existing libmcheck.a works if you link the main program with\n--export-dynamic or otherwise arrange for an exported\n__malloc_initialize_hook (even unversioned).\n\nWhat happens with my patch is that the definition has the proper symbol\nversion, so the link editor matches it against the definition in libc.so\nand notices that it needs to be listed in .dynsym.\n\nThanks,\nFlorian","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-83934-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"NIdrY0pj\"; dkim-atps=neutral","sourceware.org; auth=none","ext-mx09.extmail.prod.ext.phx2.redhat.com;\n\tdmarc=none (p=none dis=none) header.from=redhat.com","ext-mx09.extmail.prod.ext.phx2.redhat.com;\n\tspf=fail smtp.mailfrom=fweimer@redhat.com"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xjmVp4zkdz9s83\n\tfor ;\n\tFri, 1 Sep 2017 01:28:50 +1000 (AEST)","(qmail 61924 invoked by alias); 31 Aug 2017 15:28:43 -0000","(qmail 61871 invoked by uid 89); 31 Aug 2017 15:28:42 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=gevjKrm3NnlAu7ts\n\ti9KhcUZ3PffU0ZCkQ79mPW7lGD6N//mR3FeoBCtwt55ZN7R9iep2Mo7/ho6eSVYo\n\t/mKTttTaTrq8VV7evfQsGcHZlDl67XfJ8RfOx3vdMhwGE2NNR7BkZgzpXjt9lpSh\n\tDCgxSKBWuF8g9Sf1dB5tJoqDwBI=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; s=default; bh=zQGjJCW/yFcCzy0V1sRsGH\n\tEPE1s=; b=NIdrY0pjXxfZDG7eRUGAPMlPbLSMCFRPqxFLtjV/w3fzG71o0ad+3I\n\tGI6MndJ2gGNc2eKWGhRv4QLKiWu6ulk6vYOK2lbEhFeIzTI/nFFnp51OviAk5Qo2\n\tvEc3vMnGDOkqEgpzS7WhqwNG6Eddnie0JsmkLYvXhvqbbcfT5mTQk=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-1.9 required=5.0 tests=BAYES_00,\n\tRP_MATCHES_RCVD,\n\tSPF_HELO_PASS autolearn=ham version=3.3.2 spammy=lacks,\n\tH*M:cab5, H*M:5106","X-HELO":"mx1.redhat.com","DMARC-Filter":"OpenDMARC Filter v1.3.2 mx1.redhat.com 828374ACC7","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","To":"Carlos O'Donell , libc-alpha@sourceware.org","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>\n\t<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>","From":"Florian Weimer ","Message-ID":"","Date":"Thu, 31 Aug 2017 17:28:27 +0200","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.2.1","MIME-Version":"1.0","In-Reply-To":"<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>","Content-Type":"text/plain; charset=utf-8","Content-Transfer-Encoding":"7bit"}},{"id":1761129,"web_url":"http://patchwork.ozlabs.org/comment/1761129/","msgid":"<2ad943f4-1b66-5df7-b348-ab0652716470@redhat.com>","list_archive_url":null,"date":"2017-08-31T15:37:04","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":22438,"url":"http://patchwork.ozlabs.org/api/people/22438/","name":"Carlos O'Donell","email":"carlos@redhat.com"},"content":"On 08/31/2017 10:28 AM, Florian Weimer wrote:\n> On 08/31/2017 05:23 PM, Carlos O'Donell wrote:\n> \n>> We deprecated __malloc_initialize_hook in libc.so.6. There is only a compat\n>> symbol there, which can only be interposed by a definition of the same version.\n>>\n>> In libmcheck.a we have __malloc_initialize_hook (unversioned) which no longer\n>> interposes the symbol in libc.so.6 (it would have interposed the default if\n>> there was one).\n>>\n>> So what does the compat_symbol_reference create in the libmcheck.a? An undefined\n>> reference to __malloc_initialize_hook@GLIBC_2_0?\n> \n> Interposition happens only if __malloc_initialize_hook is listed in the\n> .dynsym section of the executable. At least some versions of binutils\n> will not add the symbol to the .dynsym section if it is unversioned and\n> there is a definition in a DSO which lacks a default version. Arguably\n> this is a bug in the link editor.\n> \n> The existing libmcheck.a works if you link the main program with\n> --export-dynamic or otherwise arrange for an exported\n> __malloc_initialize_hook (even unversioned).\n> \n> What happens with my patch is that the definition has the proper symbol\n> version, so the link editor matches it against the definition in libc.so\n> and notices that it needs to be listed in .dynsym.\n\nSo it makes it look like there was an old reference to\n__malloc_initialize_hook@GLIBC_2.0?\n\nIn that case I think we should not play games with the existing macro.\n\nI would like to see a new macro that does what it says, rather than use the\nexisting macro in the wrong way. Even if the new macro is just a copy.\n\nThis looks like a real problem for glibc, particularly if we need to continue\nto use, at least internally, certain old versions of symbols. So having a\nnew macro for this is fine.","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-83935-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"dBfE7BP0\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xjmhk6M9lz9sD5\n\tfor ;\n\tFri, 1 Sep 2017 01:37:26 +1000 (AEST)","(qmail 95397 invoked by alias); 31 Aug 2017 15:37:20 -0000","(qmail 95388 invoked by uid 89); 31 Aug 2017 15:37:19 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=B5qfWAidmmILUolW\n\tYxVEvDyUJ1cmyYKU2Pu6sZzJiVcgbhWB9E5R/tuMi9oMjVntkQ/Fs7TQMPjIN/Bh\n\tRVF6hR2uFs67nD9zlqRmS1eAl74yPUAGysSGHVWJssDCzBa/ESx9iwGoEbrtjnqU\n\tCuP1n3LOUE5xIJKTju07zdnpn60=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; s=default; bh=xVM3EQ/1NXzcoPDAvK2KR/\n\tw8nGw=; b=dBfE7BP0g3nU4gM3ujPbHOl1SPiJoL7bLg9iMsbOgWewhUFITcfNHF\n\tUwLzGfwTOE7wao9emBBnIHf1ks+DWku68qFiOVexuGmlfRBC9kWts4h/Hfa2jjL6\n\ternDrbV8mHf+JhqaPa/f/zmCcjJibc33bAhXdQbvFuO8v8nPDU3TQ=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-1.9 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_DNSWL_NONE,\n\tRCVD_IN_SORBS_SPAM autolearn=no version=3.3.2\n\tspammy=H*RU:209.85.216.170,\n\tHx-spam-relays-external:209.85.216.170, H*M:b348","X-HELO":"mail-qt0-f170.google.com","X-Google-DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed;\n\td=1e100.net; s=20161025;\n\th=x-gm-message-state:subject:to:references:from:organization\n\t:message-id:date:user-agent:mime-version:in-reply-to\n\t:content-language:content-transfer-encoding;\n\tbh=LIdy9Q1FK0kfDJo65qzJQUVdr39OlHuYVTduUmmDYJc=;\n\tb=FTNXdk4wmVduwjVS8VXxGY8R56gyAKwoVZkiZERfP+7Ubcus8eEwqW7Pn5kIGQe0Rm\n\tRTN7weEAMWqQ/Dy3KmACpw1KRK/j7KXU1oCk2rcS40WO1rZd+1awFl9bGLF8gTxeRlKL\n\trk3pGxOS3+RBaYKMyENFULif0PeTNr388YAFF6w2e12YI/rP5zDQEgFNFOd9T0SUlWMg\n\tjKJjyhOPkUALY8YJLHEmF8nHOI/7XL3L1D7M8dSs/QkyE3g5cMlaER1qnwMpCaoHvXgQ\n\tMcKRQooqPfXZGvNAuFqVgMinqNACOLlKvbgvcUQuU7+aIKS5RJpU19pDBtEE7aaDcwOs\n\tbQdw==","X-Gm-Message-State":"AHYfb5gr2/Btxe1MihR6hLYISlrClGtOEskstdR02wPUI5JjBIr/AZUU\n\t50xCYkm3E9FlclcvRQtiJw==","X-Google-Smtp-Source":"ADKCNb5CmpA2qni/QHpwBmfx4EOUu5Ky2Lmp2+55pV0YOuw26BgGNH7MXTJLKvmHDBv9/X5nNa/ugQ==","X-Received":"by 10.200.22.170 with SMTP id r39mr7955158qtj.61.1504193827405; \n\tThu, 31 Aug 2017 08:37:07 -0700 (PDT)","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","To":"Florian Weimer , libc-alpha@sourceware.org","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>\n\t<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>\n\t","From":"Carlos O'Donell ","Message-ID":"<2ad943f4-1b66-5df7-b348-ab0652716470@redhat.com>","Date":"Thu, 31 Aug 2017 10:37:04 -0500","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.2.1","MIME-Version":"1.0","In-Reply-To":"","Content-Type":"text/plain; charset=utf-8","Content-Transfer-Encoding":"7bit"}},{"id":1761133,"web_url":"http://patchwork.ozlabs.org/comment/1761133/","msgid":"","list_archive_url":null,"date":"2017-08-31T15:42:15","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":65878,"url":"http://patchwork.ozlabs.org/api/people/65878/","name":"Zack Weinberg","email":"zackw@panix.com"},"content":"On Thu, Aug 31, 2017 at 11:37 AM, Carlos O'Donell wrote:\n> I would like to see a new macro that does what it says, rather than use the\n> existing macro in the wrong way. Even if the new macro is just a copy.\n>\n> This looks like a real problem for glibc, particularly if we need to continue\n> to use, at least internally, certain old versions of symbols. So having a\n> new macro for this is fine.\n\nI see immediate uses for this macro in the test suite, verifying that\ncompat symbols continue to work correctly... (particularly thinking\nof the messy and totally untested old-FILE support).\n\nzw","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-83936-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"Edkz456z\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xjmpg23dVz9s83\n\tfor ;\n\tFri, 1 Sep 2017 01:42:35 +1000 (AEST)","(qmail 109751 invoked by alias); 31 Aug 2017 15:42:28 -0000","(qmail 109740 invoked by uid 89); 31 Aug 2017 15:42:28 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:mime-version:in-reply-to:references:from:date\n\t:message-id:subject:to:cc:content-type; q=dns; s=default; b=iOby\n\tMxryvIURFWNnVfQfkkpjYkvtgtjFx1Ni8aN03ZYPvIKiiUn8p0y1C5UGBy6IVMWo\n\thNbD9d6uvGdtAI1po5Gx8VofwyEklXR8c/EGU/ZgrKrzcHbibCcpVVFtj6eB7Wkw\n\tNQojGkZ5eULaJUMgSogFkRofes+cv64gT5+bYiw=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:mime-version:in-reply-to:references:from:date\n\t:message-id:subject:to:cc:content-type; s=default; bh=8DO7r9+Bfe\n\t1ww+IX2aD8f+K4Fis=; b=Edkz456z9ww+0m2pdczDBQXH1CQRn+S7j8VqGabfZV\n\tpX6XfBoJtuUcXIySnjEKxQxzj2Wl+85faEcghToLKXT9L5UdVqqd75kglEKJx6xe\n\tmlC9Y0EW0Kh9F/eYBajuag15CBI3McDSzX+L1Q8oIzgn0Uzz0N6UtEkrqzmtXQq3\n\ts=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-1.5 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_SORBS_SPAM, RP_MATCHES_RCVD,\n\tSPF_PASS autolearn=no version=3.3.2 spammy=Hx-languages-length:686","X-HELO":"mailbackend.panix.com","X-Gm-Message-State":"AHPjjUiccIPsahdah4uJ1vgOpUL9FfLuQWXJnhz0rKoK2e91Gz+QJlzq\n\tNNQeDeuilurJW+Q0Qtyfqp2kvZm9Pw==","X-Google-Smtp-Source":"ADKCNb6S6a3Z5uBAvcUBbSpPObOlqeVw0RA0lPIk6vqdCHcRBmxYrigQ560kFvSMwIPDX5CW0+U1Vs8lEBp5jGk/a7w=","X-Received":"by 10.36.64.21 with SMTP id n21mr1238826ita.135.1504194136182;\n\tThu, 31 Aug 2017 08:42:16 -0700 (PDT)","MIME-Version":"1.0","In-Reply-To":"<2ad943f4-1b66-5df7-b348-ab0652716470@redhat.com>","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>\n\t<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>\n\t\n\t<2ad943f4-1b66-5df7-b348-ab0652716470@redhat.com>","From":"Zack Weinberg ","Date":"Thu, 31 Aug 2017 11:42:15 -0400","X-Gmail-Original-Message-ID":"","Message-ID":"","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","To":"\"Carlos O'Donell\" ","Cc":"Florian Weimer ,\n\tGNU C Library ","Content-Type":"text/plain; charset=\"UTF-8\""}},{"id":1761140,"web_url":"http://patchwork.ozlabs.org/comment/1761140/","msgid":"<61a94166-1768-4014-27f7-34a9a6ff24b8@redhat.com>","list_archive_url":null,"date":"2017-08-31T16:00:15","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":14312,"url":"http://patchwork.ozlabs.org/api/people/14312/","name":"Florian Weimer","email":"fweimer@redhat.com"},"content":"On 08/31/2017 05:42 PM, Zack Weinberg wrote:\n> On Thu, Aug 31, 2017 at 11:37 AM, Carlos O'Donell wrote:\n>> I would like to see a new macro that does what it says, rather than use the\n>> existing macro in the wrong way. Even if the new macro is just a copy.\n>>\n>> This looks like a real problem for glibc, particularly if we need to continue\n>> to use, at least internally, certain old versions of symbols. So having a\n>> new macro for this is fine.\n> \n> I see immediate uses for this macro in the test suite, verifying that\n> compat symbols continue to work correctly... (particularly thinking\n> of the messy and totally untested old-FILE support).\n\nThat's the exact purpose of compat_symbol_reference. I think Carlos is\nobjecting to its use for a *definition*.\n\nThanks,\nFlorian","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-83938-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"V7F19yTR\"; dkim-atps=neutral","sourceware.org; auth=none","ext-mx08.extmail.prod.ext.phx2.redhat.com;\n\tdmarc=none (p=none dis=none) header.from=redhat.com","ext-mx08.extmail.prod.ext.phx2.redhat.com;\n\tspf=fail smtp.mailfrom=fweimer@redhat.com"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xjnCV44WNz9sD9\n\tfor ;\n\tFri, 1 Sep 2017 02:00:38 +1000 (AEST)","(qmail 2178 invoked by alias); 31 Aug 2017 16:00:32 -0000","(qmail 2169 invoked by uid 89); 31 Aug 2017 16:00:32 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:cc:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=c573ZcV4asKxT2sx\n\tVChA/UO3fT4uFEnVuv7xhSLgthB150jS1oZBaGf5tulSI78ELhIZfaJQOsLN/LVs\n\tk4ko2WbzLHVxvs3K4qs6J57oIEfrj8eaVGFVmeKzh9tmZvX/nx8awr7Fv7QIaQjD\n\tlEy2qadzAjwH5l6BoF/pzSt28XQ=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:cc:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; s=default; bh=iB/FpJFZ1bAeUhHSqK+ALb\n\toTw00=; b=V7F19yTRzPUA3xXZPU/SJhd4EI/b3GfI7UeOKkzXQsgBolPUjhVTKZ\n\t4lOrgLONlKNuz5VzRm89xhomCHCoGw5lq8254gK9DAQVK0i7hZZJwbqgmYaPVt7n\n\tvFE2nMIrQ8vg65KP4qDRcgFnkXX9tYHA6sxPmMy7rYyaa3Ynw3uyA=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-1.9 required=5.0 tests=BAYES_00,\n\tRP_MATCHES_RCVD,\n\tSPF_HELO_PASS autolearn=ham version=3.3.2 spammy=","X-HELO":"mx1.redhat.com","DMARC-Filter":"OpenDMARC Filter v1.3.2 mx1.redhat.com B9D99C05681B","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","To":"Zack Weinberg , Carlos O'Donell ","Cc":"GNU C Library ","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>\n\t<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>\n\t\n\t<2ad943f4-1b66-5df7-b348-ab0652716470@redhat.com>\n\t","From":"Florian Weimer ","Message-ID":"<61a94166-1768-4014-27f7-34a9a6ff24b8@redhat.com>","Date":"Thu, 31 Aug 2017 18:00:15 +0200","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.2.1","MIME-Version":"1.0","In-Reply-To":"","Content-Type":"text/plain; charset=utf-8","Content-Transfer-Encoding":"7bit"}},{"id":1761143,"web_url":"http://patchwork.ozlabs.org/comment/1761143/","msgid":"","list_archive_url":null,"date":"2017-08-31T16:02:52","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":4349,"url":"http://patchwork.ozlabs.org/api/people/4349/","name":"Joseph Myers","email":"joseph@codesourcery.com"},"content":"On Thu, 31 Aug 2017, Florian Weimer wrote:\n\n> Interposition happens only if __malloc_initialize_hook is listed in the\n> .dynsym section of the executable. At least some versions of binutils\n> will not add the symbol to the .dynsym section if it is unversioned and\n> there is a definition in a DSO which lacks a default version. Arguably\n> this is a bug in the link editor.\n\nSounds like a feature to me, not a bug. After all, it's essentially what \nmakes it possible for us to obsolete (stop working for new programs as far \nas possible) symbols whose interface is that the program interposes them, \nsuch as matherr. Effectively, a DSO definition that lacks a default \nversion should be treated exactly the same at static link time as no DSO \ndefinition at all.","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-83940-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"m8Mr0rVY\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xjnGr5zjdz9sMN\n\tfor ;\n\tFri, 1 Sep 2017 02:03:17 +1000 (AEST)","(qmail 21430 invoked by alias); 31 Aug 2017 16:03:12 -0000","(qmail 21380 invoked by uid 89); 31 Aug 2017 16:03:11 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; q=dns; s=default; b=weH7/\n\tAsmYWC2+D0Zauw8s9Y7npOxAHRuQJP4wR9KlKXCg52NKRVlWFq8DoH4WuBBot/ao\n\t7biO9sD+KWUVPO5HuXLhJ8oYvmbS1qdew9Mmbw3RcxE8ZFTtI7haUoJY1DOXRoFA\n\tSZrljb0tfXif6VSr1mO+tMM7orbCcbQxxm4/cw=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; s=default; bh=UPH/huhpruC\n\txbbNwEPBHrEHvbs8=; b=m8Mr0rVYWwOHVuYvrVqYLkRxgoYBMmHKg0Nys7tjUwN\n\tCMq2V1U/PlCRNkqXCIFWsseaJtbgapCqCqjYzdmNzS6VkWura59EeG4q6zyQzk3h\n\tIS+gEIL1NEz1/DldJGCaJ0fT3EF8J2DnrcPg5bKAkzX663pUFicC7icRDFx0Jozc\n\t=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-2.0 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_DNSWL_NONE, SPF_PASS,\n\tURIBL_RED autolearn=ham version=3.3.2 spammy=","X-HELO":"relay1.mentorg.com","Date":"Thu, 31 Aug 2017 16:02:52 +0000","From":"Joseph Myers ","To":"Florian Weimer ","CC":"Carlos O'Donell , ","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","In-Reply-To":"","Message-ID":"","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>\n\t<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>\n\t","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\n\tsvr-ies-mbx-01.mgc.mentorg.com (139.181.222.1)"}},{"id":1761145,"web_url":"http://patchwork.ozlabs.org/comment/1761145/","msgid":"","list_archive_url":null,"date":"2017-08-31T16:04:34","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":4349,"url":"http://patchwork.ozlabs.org/api/people/4349/","name":"Joseph Myers","email":"joseph@codesourcery.com"},"content":"On Thu, 31 Aug 2017, Florian Weimer wrote:\n\n> On 08/31/2017 05:42 PM, Zack Weinberg wrote:\n> > On Thu, Aug 31, 2017 at 11:37 AM, Carlos O'Donell wrote:\n> >> I would like to see a new macro that does what it says, rather than use the\n> >> existing macro in the wrong way. Even if the new macro is just a copy.\n> >>\n> >> This looks like a real problem for glibc, particularly if we need to continue\n> >> to use, at least internally, certain old versions of symbols. So having a\n> >> new macro for this is fine.\n> > \n> > I see immediate uses for this macro in the test suite, verifying that\n> > compat symbols continue to work correctly... (particularly thinking\n> > of the messy and totally untested old-FILE support).\n> \n> That's the exact purpose of compat_symbol_reference. I think Carlos is\n> objecting to its use for a *definition*.\n\nWell, I used it for the definitions of matherr and _LIB_VERSION in my \ntests of those compat symbols, because it does exactly what's expected: \nmakes the definitions in the tests refer to the same entity as the compat \nsymbols in the shared libraries, rather than being completely independent \nentities as they would by default.","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-83941-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"YKP08KtO\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xjnJX6LlYz9sNr\n\tfor ;\n\tFri, 1 Sep 2017 02:05:00 +1000 (AEST)","(qmail 47901 invoked by alias); 31 Aug 2017 16:04:53 -0000","(qmail 47469 invoked by uid 89); 31 Aug 2017 16:04:53 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; q=dns; s=default; b=l2lg2\n\tySqPLW9nqcmZ+bCvryvL3Onju1VoPgUM1STXDwnNXv5xAc2NO1fc6Xsc67rLmYEM\n\t8pHYeLIQGkO2zEBVql/ZWjRMry5rW7LmvR6kfuNomntsaJl44pYdE1oikp5r9JoY\n\tnLikf92pngw5CUVoqMhsgaPBzkVs5IWd6iSE5s=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; s=default; bh=1xnoaOsMTS4\n\tpmlXgm27eh71Iqes=; b=YKP08KtORQ4omLBBn0MF7ZIgCEXpJeXq9sv3cn7r/94\n\t1KkwNtWiZCd0kmHD5rEQ+GT/Rtp0mwOlOeD/jR5SDHIvKun5EkHaEZorqGBe6mEo\n\taxdDsDCFUUpEGUvdNIw0a+qTEdAOYlLrz4ymPIWFixWUJe0rMWCy99XhFsN7G/+4\n\t=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-2.0 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_DNSWL_NONE, SPF_PASS,\n\tURIBL_RED autolearn=ham version=3.3.2 spammy=","X-HELO":"relay1.mentorg.com","Date":"Thu, 31 Aug 2017 16:04:34 +0000","From":"Joseph Myers ","To":"Florian Weimer ","CC":"Zack Weinberg , Carlos O'Donell , GNU\n\tC Library ","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","In-Reply-To":"<61a94166-1768-4014-27f7-34a9a6ff24b8@redhat.com>","Message-ID":"","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>\n\t<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>\n\t\n\t<2ad943f4-1b66-5df7-b348-ab0652716470@redhat.com>\n\t\n\t<61a94166-1768-4014-27f7-34a9a6ff24b8@redhat.com>","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\n\tsvr-ies-mbx-01.mgc.mentorg.com (139.181.222.1)"}},{"id":1761148,"web_url":"http://patchwork.ozlabs.org/comment/1761148/","msgid":"","list_archive_url":null,"date":"2017-08-31T16:08:32","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":22438,"url":"http://patchwork.ozlabs.org/api/people/22438/","name":"Carlos O'Donell","email":"carlos@redhat.com"},"content":"On 08/31/2017 11:04 AM, Joseph Myers wrote:\n> On Thu, 31 Aug 2017, Florian Weimer wrote:\n> \n>> On 08/31/2017 05:42 PM, Zack Weinberg wrote:\n>>> On Thu, Aug 31, 2017 at 11:37 AM, Carlos O'Donell wrote:\n>>>> I would like to see a new macro that does what it says, rather than use the\n>>>> existing macro in the wrong way. Even if the new macro is just a copy.\n>>>>\n>>>> This looks like a real problem for glibc, particularly if we need to continue\n>>>> to use, at least internally, certain old versions of symbols. So having a\n>>>> new macro for this is fine.\n>>>\n>>> I see immediate uses for this macro in the test suite, verifying that\n>>> compat symbols continue to work correctly... (particularly thinking\n>>> of the messy and totally untested old-FILE support).\n>>\n>> That's the exact purpose of compat_symbol_reference. I think Carlos is\n>> objecting to its use for a *definition*.\n> \n> Well, I used it for the definitions of matherr and _LIB_VERSION in my \n> tests of those compat symbols, because it does exactly what's expected: \n> makes the definitions in the tests refer to the same entity as the compat \n> symbols in the shared libraries, rather than being completely independent \n> entities as they would by default.\n \nWhile it does what's expected, the macro API wasn't designed with that in\nmind was it? I am objecting to using a macro not designed for this\npurpose, and suggesting a new macro that makes the intent clear.","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-83942-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"RDnUZpRG\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xjnP04vsxz9sD9\n\tfor ;\n\tFri, 1 Sep 2017 02:08:52 +1000 (AEST)","(qmail 71538 invoked by alias); 31 Aug 2017 16:08:47 -0000","(qmail 71518 invoked by uid 89); 31 Aug 2017 16:08:47 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:cc:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=IKeIg0utt+FawF0a\n\tYV4eio3rzK3RGfpNTxDN61AyqCdWiUeMPqmvTupToQGa70MXBIOJFkgEK11YVxSY\n\tyacTNn3aKjdNiH8lm0X5gJQZ9GXL9kVj7oHWjzb3Nh/DpKui+EU33h0eezhIgapx\n\tEz6axdc/F5Egjx36RvzXE4ZwfXQ=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:cc:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; s=default; bh=hWfh/jq5eRcBLE7wX4qSRY\n\tBnbBI=; b=RDnUZpRGuHDNTuVmhSSxbSzNQq2x4Zd1f6BZ8aMySJnWpHpYdPvYot\n\tht81YW810/JYUNMVpNqHzm/9PWsNWzXh6KMmmjmptbR10qBATwy5AbGvxLQYotX/\n\tuj3bciplS2yvip6yIjZ4LcPZf7lV1CeA4HjHgB8NMlFx16V8kz9BE=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-1.8 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_DNSWL_NONE,\n\tRCVD_IN_SORBS_SPAM autolearn=no version=3.3.2 spammy=H*M:5435,\n\tHx-languages-length:1543, H*M:af58","X-HELO":"mail-qk0-f179.google.com","X-Google-DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed;\n\td=1e100.net; s=20161025;\n\th=x-gm-message-state:subject:to:cc:references:from:organization\n\t:message-id:date:user-agent:mime-version:in-reply-to\n\t:content-language:content-transfer-encoding;\n\tbh=meMR6sPmtV3l/aPwJoS2wVqq1HxAEvcQeDp9tp14zuU=;\n\tb=gBadgfLIcmBqcX2vR0xm6kEfparkQS7vAPzKgufhA2nP815z9480aVinvFmNo+wfcV\n\temLKIJREyMuyeRke93Xru9wX/M/dim7xzpJYsesAhDcJkLX5lbrdWEGeNcx6EgG5B6eq\n\t3kX/U2Ks+Dimu0w3q+UEuxt5dwGUD+hhR5+BuITLKeSUSd+epm6EZcQPiGhQVweu5H2z\n\tLzw2XoOBJpLBArmtoFcwlIWLOXOpUW2GcIolvRepaMJ2PblNhsJKWqivTh9PEYHRUCO4\n\tKgWabSGQ+RXkG9UiM9tn0N0R3NBbgAiFXSA4p6YEKJdPAczD5ey0OiQ0o61OtJfIMUAm\n\tTOZg==","X-Gm-Message-State":"AHYfb5jGPhcdABLSfoJ6y+OCxYbH5mbTfbwEkkcZ2Phfr3kvRnocgPeD\n\tiK2RkoaGe3YDUjeXtpcgkw==","X-Google-Smtp-Source":"ADKCNb4MBUXa9SceEARippPbMwnyFHgdgimkM4YIfjICQ1wV8T3xEadkGMbDwQRlM66KQilRgiagSA==","X-Received":"by 10.55.204.137 with SMTP id n9mr4325198qkl.125.1504195715017; \n\tThu, 31 Aug 2017 09:08:35 -0700 (PDT)","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","To":"Joseph Myers ,\n\tFlorian Weimer ","Cc":"Zack Weinberg ,\n\tGNU C Library ","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>\n\t<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>\n\t\n\t<2ad943f4-1b66-5df7-b348-ab0652716470@redhat.com>\n\t\n\t<61a94166-1768-4014-27f7-34a9a6ff24b8@redhat.com>\n\t","From":"Carlos O'Donell ","Message-ID":"","Date":"Thu, 31 Aug 2017 11:08:32 -0500","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.2.1","MIME-Version":"1.0","In-Reply-To":"","Content-Type":"text/plain; charset=utf-8","Content-Transfer-Encoding":"7bit"}},{"id":1761156,"web_url":"http://patchwork.ozlabs.org/comment/1761156/","msgid":"","list_archive_url":null,"date":"2017-08-31T16:17:56","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":4349,"url":"http://patchwork.ozlabs.org/api/people/4349/","name":"Joseph Myers","email":"joseph@codesourcery.com"},"content":"On Thu, 31 Aug 2017, Carlos O'Donell wrote:\n\n> >> That's the exact purpose of compat_symbol_reference. I think Carlos is\n> >> objecting to its use for a *definition*.\n> > \n> > Well, I used it for the definitions of matherr and _LIB_VERSION in my \n> > tests of those compat symbols, because it does exactly what's expected: \n> > makes the definitions in the tests refer to the same entity as the compat \n> > symbols in the shared libraries, rather than being completely independent \n> > entities as they would by default.\n> \n> While it does what's expected, the macro API wasn't designed with that in\n> mind was it? I am objecting to using a macro not designed for this\n> purpose, and suggesting a new macro that makes the intent clear.\n\nI think \"reference\" naturally covers both cases of making the symbol refer \nto the same entity in both places.","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-83946-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"NmSZJBnp\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xjnbw07bSz9sPm\n\tfor ;\n\tFri, 1 Sep 2017 02:18:19 +1000 (AEST)","(qmail 129220 invoked by alias); 31 Aug 2017 16:18:14 -0000","(qmail 129204 invoked by uid 89); 31 Aug 2017 16:18:14 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; q=dns; s=default; b=NuFye\n\t+7Y2Dx3uMgcgXk9Q2nrcZrWEwjNnUeHsVDo1x9sfmf/kbpIqAiRc7EiESIN9jJOg\n\tI5IX30HUGqt4biVW7yc1XN6/9F8bgliH5TooC5Ypd82nOI2Td08NGCNoHkC91Gan\n\tC2ABFRVURa8+xebxy3zRxoj6WFF6VtP9+BwABM=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; s=default; bh=BGe7cyWBiAx\n\tY1sFgOK4u4zzqC+Y=; b=NmSZJBnpU8b920OUa078rNa7Vk4RNTZ50vsXgEaYYYy\n\tIjCFgU8H5xTmS83M54b97eHJoCgUrXXH2xeWxeCz3H3JRCLxlR0XkNEl/YVYbeux\n\t028kHOI3Kb8pQLYyqYE8qHUARNJIKiQ5zLjCV9Xk9CEbDEAY5zXU4XTmjiclx6AU\n\t=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-2.0 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_DNSWL_NONE, SPF_PASS,\n\tURIBL_RED autolearn=ham version=3.3.2 spammy=","X-HELO":"relay1.mentorg.com","Date":"Thu, 31 Aug 2017 16:17:56 +0000","From":"Joseph Myers ","To":"Carlos O'Donell ","CC":"Florian Weimer , Zack Weinberg , GNU\n\tC Library ","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","In-Reply-To":"","Message-ID":"","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>\n\t<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>\n\t\n\t<2ad943f4-1b66-5df7-b348-ab0652716470@redhat.com>\n\t\n\t<61a94166-1768-4014-27f7-34a9a6ff24b8@redhat.com>\n\t\n\t","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\n\tsvr-ies-mbx-01.mgc.mentorg.com (139.181.222.1)"}},{"id":1761173,"web_url":"http://patchwork.ozlabs.org/comment/1761173/","msgid":"","list_archive_url":null,"date":"2017-08-31T16:34:32","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":14312,"url":"http://patchwork.ozlabs.org/api/people/14312/","name":"Florian Weimer","email":"fweimer@redhat.com"},"content":"On 08/31/2017 06:02 PM, Joseph Myers wrote:\n> On Thu, 31 Aug 2017, Florian Weimer wrote:\n> \n>> Interposition happens only if __malloc_initialize_hook is listed in the\n>> .dynsym section of the executable. At least some versions of binutils\n>> will not add the symbol to the .dynsym section if it is unversioned and\n>> there is a definition in a DSO which lacks a default version. Arguably\n>> this is a bug in the link editor.\n> \n> Sounds like a feature to me, not a bug. After all, it's essentially what \n> makes it possible for us to obsolete (stop working for new programs as far \n> as possible) symbols whose interface is that the program interposes them, \n> such as matherr. Effectively, a DSO definition that lacks a default \n> version should be treated exactly the same at static link time as no DSO \n> definition at all.\n\nBut that's not what happens if you compile with --export-dynamic. That\n--export-dynamic changes behavior here is very surprising. That's why I\nthink this could be considered a bug in the link editor.\n\nThanks,\nFlorian","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-83949-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"WL6FH1oJ\"; dkim-atps=neutral","sourceware.org; auth=none","ext-mx07.extmail.prod.ext.phx2.redhat.com;\n\tdmarc=none (p=none dis=none) header.from=redhat.com","ext-mx07.extmail.prod.ext.phx2.redhat.com;\n\tspf=fail smtp.mailfrom=fweimer@redhat.com"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xjnz21fdLz9sPt\n\tfor ;\n\tFri, 1 Sep 2017 02:34:54 +1000 (AEST)","(qmail 15240 invoked by alias); 31 Aug 2017 16:34:48 -0000","(qmail 15190 invoked by uid 89); 31 Aug 2017 16:34:48 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:cc:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=ijtaMhrsBrkcPUC0\n\tsrJ6tHXng9omMUJSlw/M2Iz5qfPF04bVvh8XptJo8PvUVashODeyR6YpJyDNrTBx\n\tw1W+QQ/EbgM6Lb4PMIGU2TXblfqOXZTQMntRObgornjgO3jMd1KgyiajIRfictWA\n\tFVlDVPXgoeeFosPSYNNy1DTZ/fM=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:cc:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; s=default; bh=uVlpIcniHc77CNVokIs52b\n\t+355A=; b=WL6FH1oJRaLIFUrTM6JXvItk9Gfoz5355NgOXz4Km/p8CHwvzWANAN\n\tB741p3n8iqNl0FogQ60uP4eXpU86ANNy67PxEMNOMZZZB8ZzlW5twygDL/5OR1Tl\n\tlULv4/oL2h0akJ50LOUDUfWBHzXceVVI7NXzetI241XxdTie0Tzis=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-0.9 required=5.0 tests=BAYES_00,\n\tKAM_LAZY_DOMAIN_SECURITY, RP_MATCHES_RCVD,\n\tSPF_HELO_PASS autolearn=no version=3.3.2 spammy=","X-HELO":"mx1.redhat.com","DMARC-Filter":"OpenDMARC Filter v1.3.2 mx1.redhat.com AD8C5C047B8D","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","To":"Joseph Myers ","Cc":"Carlos O'Donell , libc-alpha@sourceware.org","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>\n\t<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>\n\t\n\t","From":"Florian Weimer ","Message-ID":"","Date":"Thu, 31 Aug 2017 18:34:32 +0200","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.2.1","MIME-Version":"1.0","In-Reply-To":"","Content-Type":"text/plain; charset=utf-8","Content-Transfer-Encoding":"7bit"}},{"id":1761184,"web_url":"http://patchwork.ozlabs.org/comment/1761184/","msgid":"","list_archive_url":null,"date":"2017-08-31T16:45:22","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":4349,"url":"http://patchwork.ozlabs.org/api/people/4349/","name":"Joseph Myers","email":"joseph@codesourcery.com"},"content":"On Thu, 31 Aug 2017, Florian Weimer wrote:\n\n> On 08/31/2017 06:02 PM, Joseph Myers wrote:\n> > On Thu, 31 Aug 2017, Florian Weimer wrote:\n> > \n> >> Interposition happens only if __malloc_initialize_hook is listed in the\n> >> .dynsym section of the executable. At least some versions of binutils\n> >> will not add the symbol to the .dynsym section if it is unversioned and\n> >> there is a definition in a DSO which lacks a default version. Arguably\n> >> this is a bug in the link editor.\n> > \n> > Sounds like a feature to me, not a bug. After all, it's essentially what \n> > makes it possible for us to obsolete (stop working for new programs as far \n> > as possible) symbols whose interface is that the program interposes them, \n> > such as matherr. Effectively, a DSO definition that lacks a default \n> > version should be treated exactly the same at static link time as no DSO \n> > definition at all.\n> \n> But that's not what happens if you compile with --export-dynamic. That\n> --export-dynamic changes behavior here is very surprising. That's why I\n> think this could be considered a bug in the link editor.\n\nThat --export-dynamic changes things sounds like a consequence of the rule \nthat an unversioned symbol can be linked to the base symbol version, \nbecause symbol versioning support wasn't added until glibc 2.1 and \nbinaries with unversioned references to glibc 2.0 symbols needed to be \nsupported. I'm not sure how to fix that (stop --export-dynamic allowing \nsymbols to link to compat symbols) - maybe that special rule should be \ndisabled for glibc symbols if a program has any dependence on a glibc \nversion greater than 2.0, is there a way we can make sure all newly linked \nprograms have such dependence to stop that rule applying to them? (If \n--export-dynamic results in *versioned* exports from the executable for \nsuch symbols as matherr or __malloc_initialize_hook, that sounds like a \nbug in ld.)","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-83951-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"ZARkAfBN\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xjpCb6hgQz9s81\n\tfor ;\n\tFri, 1 Sep 2017 02:45:47 +1000 (AEST)","(qmail 30486 invoked by alias); 31 Aug 2017 16:45:41 -0000","(qmail 30434 invoked by uid 89); 31 Aug 2017 16:45:40 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; q=dns; s=default; b=XMjJg\n\thXQKf5ebU12hAC/UajWfcUZafvn3TG0DT6O65lGqTo2wKCM7kfTjEJ0gHC9MY25W\n\tF1sccf4N2magblC98ZbiP66ASlyJmogUABCjDilOvQe4hF0pM9cJPbAaxynwHd5W\n\tflauACilj9HNnn90x7xNyL+VjEcZs7nKqNzXSs=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:date:from:to:cc:subject:in-reply-to:message-id\n\t:references:mime-version:content-type; s=default; bh=lFVuxVOgUXH\n\tVZcfrgRA69fGsFNY=; b=ZARkAfBNcPDFgp2P77ObSWcKXXquO/8V0BQmDhXIoah\n\tfODaldR+6a0lWjmePM7IkcXfhCPekou00cqFOT9ye+7K6qUdfyjFKjMDIyhri0M4\n\tYoSbhYOXbwUF5W/VSha+x3Z6Hqzmideal9bC4+FYO9sv0v10l6fWVKAU0ZedClTI\n\t=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-2.0 required=5.0 tests=AWL, BAYES_00,\n\tRCVD_IN_DNSWL_NONE, SPF_PASS,\n\tURIBL_RED autolearn=ham version=3.3.2 spammy=","X-HELO":"relay1.mentorg.com","Date":"Thu, 31 Aug 2017 16:45:22 +0000","From":"Joseph Myers ","To":"Florian Weimer ","CC":"Carlos O'Donell , ","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","In-Reply-To":"","Message-ID":"","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>\n\t<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>\n\t\n\t\n\t","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\n\tsvr-ies-mbx-01.mgc.mentorg.com (139.181.222.1)"}},{"id":1770199,"web_url":"http://patchwork.ozlabs.org/comment/1770199/","msgid":"<5bf3d95b-cdfb-c55f-478f-be61baa846d3@redhat.com>","list_archive_url":null,"date":"2017-09-18T14:24:47","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":14312,"url":"http://patchwork.ozlabs.org/api/people/14312/","name":"Florian Weimer","email":"fweimer@redhat.com"},"content":"On 08/31/2017 06:08 PM, Carlos O'Donell wrote:\n> On 08/31/2017 11:04 AM, Joseph Myers wrote:\n>> On Thu, 31 Aug 2017, Florian Weimer wrote:\n>>\n>>> On 08/31/2017 05:42 PM, Zack Weinberg wrote:\n>>>> On Thu, Aug 31, 2017 at 11:37 AM, Carlos O'Donell wrote:\n>>>>> I would like to see a new macro that does what it says, rather than use the\n>>>>> existing macro in the wrong way. Even if the new macro is just a copy.\n>>>>>\n>>>>> This looks like a real problem for glibc, particularly if we need to continue\n>>>>> to use, at least internally, certain old versions of symbols. So having a\n>>>>> new macro for this is fine.\n>>>>\n>>>> I see immediate uses for this macro in the test suite, verifying that\n>>>> compat symbols continue to work correctly... (particularly thinking\n>>>> of the messy and totally untested old-FILE support).\n>>>\n>>> That's the exact purpose of compat_symbol_reference. I think Carlos is\n>>> objecting to its use for a *definition*.\n>>\n>> Well, I used it for the definitions of matherr and _LIB_VERSION in my\n>> tests of those compat symbols, because it does exactly what's expected:\n>> makes the definitions in the tests refer to the same entity as the compat\n>> symbols in the shared libraries, rather than being completely independent\n>> entities as they would by default.\n> \n> While it does what's expected, the macro API wasn't designed with that in\n> mind was it?\n\nI would have used it in tst-mallocstate for __malloc_initialize_hook if \nI had understood what was going on. I think the usages are so similar \nthat we do not need a new macro.\n\nThanks,\nFlorian","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-84681-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"hbl2y7l0\"; dkim-atps=neutral","sourceware.org; auth=none","ext-mx06.extmail.prod.ext.phx2.redhat.com;\n\tdmarc=none (p=none dis=none) header.from=redhat.com","ext-mx06.extmail.prod.ext.phx2.redhat.com;\n\tspf=fail smtp.mailfrom=fweimer@redhat.com"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3xwpDt51Thz9s7M\n\tfor ;\n\tTue, 19 Sep 2017 00:25:02 +1000 (AEST)","(qmail 97711 invoked by alias); 18 Sep 2017 14:24:55 -0000","(qmail 96794 invoked by uid 89); 18 Sep 2017 14:24:55 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:cc:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=tmXj/J7Dt1JaxNZw\n\tv1jXsN9zz/DoCaS73g009/e8vdN/jC7TFhvx7L8e7GF7DJaBEZr8UQ+et7EGGK3A\n\tRnTwuhGoAFdZKy4zB6TFpoGwJyictNoj9G464lkejRU8Zd7OHjeldU2k/YWhXJMm\n\t5q/dziCTVsLh09ddpoyAoSk1Czk=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:cc:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; s=default; bh=avdkeCiYkTWyWBNGoK03Fx\n\tUVeQA=; b=hbl2y7l0q5y3iS8qpKAvsPA3XF4pVu5eAgwRkncPNTar1jfl6QWtnd\n\t3qMLGl13IDOm3HARrDuQG9QMutFZSlsm0VO7dZEpmbBgWcYwFDD8EN/eu2a/+c8z\n\tW5rupcQdP6npQbFtzpbXVuFc+BhixQ5gnBN7TdO1oXkmaFQZ19kqo=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-1.9 required=5.0 tests=BAYES_00,\n\tRP_MATCHES_RCVD,\n\tSPF_HELO_PASS autolearn=ham version=3.3.2 spammy=ODonell,\n\todonell, HTo:U*carlos","X-HELO":"mx1.redhat.com","DMARC-Filter":"OpenDMARC Filter v1.3.2 mx1.redhat.com 051EA356FC","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","To":"Carlos O'Donell ,\n\tJoseph Myers ","Cc":"Zack Weinberg ,\n\tGNU C Library ","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>\n\t<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>\n\t\n\t<2ad943f4-1b66-5df7-b348-ab0652716470@redhat.com>\n\t\n\t<61a94166-1768-4014-27f7-34a9a6ff24b8@redhat.com>\n\t\n\t","From":"Florian Weimer ","Message-ID":"<5bf3d95b-cdfb-c55f-478f-be61baa846d3@redhat.com>","Date":"Mon, 18 Sep 2017 16:24:47 +0200","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.3.0","MIME-Version":"1.0","In-Reply-To":"","Content-Type":"text/plain; charset=utf-8; format=flowed","Content-Transfer-Encoding":"7bit"}},{"id":1780452,"web_url":"http://patchwork.ozlabs.org/comment/1780452/","msgid":"<834aa29d-d8ad-b516-ab12-98b99ad4faf5@redhat.com>","list_archive_url":null,"date":"2017-10-05T10:45:56","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":14312,"url":"http://patchwork.ozlabs.org/api/people/14312/","name":"Florian Weimer","email":"fweimer@redhat.com"},"content":"On 09/18/2017 04:24 PM, Florian Weimer wrote:\n> On 08/31/2017 06:08 PM, Carlos O'Donell wrote:\n>> On 08/31/2017 11:04 AM, Joseph Myers wrote:\n>>> On Thu, 31 Aug 2017, Florian Weimer wrote:\n>>>\n>>>> On 08/31/2017 05:42 PM, Zack Weinberg wrote:\n>>>>> On Thu, Aug 31, 2017 at 11:37 AM, Carlos O'Donell \n>>>>> wrote:\n>>>>>> I would like to see a new macro that does what it says, rather \n>>>>>> than use the\n>>>>>> existing macro in the wrong way. Even if the new macro is just a \n>>>>>> copy.\n>>>>>>\n>>>>>> This looks like a real problem for glibc, particularly if we need \n>>>>>> to continue\n>>>>>> to use, at least internally, certain old versions of symbols. So \n>>>>>> having a\n>>>>>> new macro for this is fine.\n>>>>>\n>>>>> I see immediate uses for this macro in the test suite, verifying that\n>>>>> compat symbols continue to work correctly...  (particularly thinking\n>>>>> of the messy and totally untested old-FILE support).\n>>>>\n>>>> That's the exact purpose of compat_symbol_reference.  I think Carlos is\n>>>> objecting to its use for a *definition*.\n>>>\n>>> Well, I used it for the definitions of matherr and _LIB_VERSION in my\n>>> tests of those compat symbols, because it does exactly what's expected:\n>>> makes the definitions in the tests refer to the same entity as the \n>>> compat\n>>> symbols in the shared libraries, rather than being completely \n>>> independent\n>>> entities as they would by default.\n>> While it does what's expected, the macro API wasn't designed with that in\n>> mind was it?\n> \n> I would have used it in tst-mallocstate for __malloc_initialize_hook if \n> I had understood what was going on.  I think the usages are so similar \n> that we do not need a new macro.\n\nCarlos, do you still have objections to the patch as posted?\n\n \n\nI do not think we need another macro with a different name for this.\n\nThanks,\nFlorian","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-85409-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"NkKHUUp0\"; dkim-atps=neutral","sourceware.org; auth=none","ext-mx01.extmail.prod.ext.phx2.redhat.com;\n\tdmarc=none (p=none dis=none) header.from=redhat.com","ext-mx01.extmail.prod.ext.phx2.redhat.com;\n\tspf=fail smtp.mailfrom=fweimer@redhat.com"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3y78Zd20gBz9t43\n\tfor ;\n\tThu, 5 Oct 2017 21:46:17 +1100 (AEDT)","(qmail 83756 invoked by alias); 5 Oct 2017 10:46:06 -0000","(qmail 83357 invoked by uid 89); 5 Oct 2017 10:46:05 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:from:to:cc:references:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=t9yLlovxZ1SutMPk\n\tiji6hw+lw57N+cceb+cpWFhpGkYfjudA8FjmDptN4BvZgQh2HZFs/SXXN4AHPHkl\n\thvhW6x9mARHsFur3Oixq0QkDzsz9nutbllLnCrEGFuon0DEgmP7EQfEXNicDFRit\n\tp7j6quzz92PBiwCbiRcmwtw0m3U=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:from:to:cc:references:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; s=default; bh=PDkMoGCM0BhJSuuOLLJy/D\n\t4wot0=; b=NkKHUUp0MojQJesKoV/WHoE9vwp146/D6kkShugwuNwAx9zGYc57YP\n\tSVvFfKFJnkFaYF65GifT44aow981IKoriD5EAg3oMWNp/qGtO/svjEVlqnu+dfRm\n\tHnlwGphmTihMx4amebValn8aQs01ZoSSfx5LiXYuSvswi5y2NWkxY=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-0.9 required=5.0 tests=BAYES_00,\n\tKAM_LAZY_DOMAIN_SECURITY, RP_MATCHES_RCVD,\n\tSPF_HELO_PASS autolearn=no version=3.3.2 spammy=on,\n\tHx-languages-length:1993","X-HELO":"mx1.redhat.com","DMARC-Filter":"OpenDMARC Filter v1.3.2 mx1.redhat.com 73CA580E6A","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","From":"Florian Weimer ","To":"Carlos O'Donell ,\n\tJoseph Myers ","Cc":"Zack Weinberg ,\n\tGNU C Library ","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>\n\t<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>\n\t\n\t<2ad943f4-1b66-5df7-b348-ab0652716470@redhat.com>\n\t\n\t<61a94166-1768-4014-27f7-34a9a6ff24b8@redhat.com>\n\t\n\t\n\t<5bf3d95b-cdfb-c55f-478f-be61baa846d3@redhat.com>","Message-ID":"<834aa29d-d8ad-b516-ab12-98b99ad4faf5@redhat.com>","Date":"Thu, 5 Oct 2017 12:45:56 +0200","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.3.0","MIME-Version":"1.0","In-Reply-To":"<5bf3d95b-cdfb-c55f-478f-be61baa846d3@redhat.com>","Content-Type":"text/plain; charset=utf-8; format=flowed","Content-Transfer-Encoding":"8bit"}},{"id":1784819,"web_url":"http://patchwork.ozlabs.org/comment/1784819/","msgid":"<60d34595-b3a7-c170-d325-59b1234e3479@redhat.com>","list_archive_url":null,"date":"2017-10-11T17:00:27","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":22438,"url":"http://patchwork.ozlabs.org/api/people/22438/","name":"Carlos O'Donell","email":"carlos@redhat.com"},"content":"On 10/05/2017 03:45 AM, Florian Weimer wrote:\n> On 09/18/2017 04:24 PM, Florian Weimer wrote:\n>> On 08/31/2017 06:08 PM, Carlos O'Donell wrote:\n>>> On 08/31/2017 11:04 AM, Joseph Myers wrote:\n>>>> On Thu, 31 Aug 2017, Florian Weimer wrote:\n>>>>\n>>>>> On 08/31/2017 05:42 PM, Zack Weinberg wrote:\n>>>>>> On Thu, Aug 31, 2017 at 11:37 AM, Carlos O'Donell wrote:\n>>>>>>> I would like to see a new macro that does what it says, rather than use the\n>>>>>>> existing macro in the wrong way. Even if the new macro is just a copy.\n>>>>>>>\n>>>>>>> This looks like a real problem for glibc, particularly if we need to continue\n>>>>>>> to use, at least internally, certain old versions of symbols. So having a\n>>>>>>> new macro for this is fine.\n>>>>>>\n>>>>>> I see immediate uses for this macro in the test suite, verifying that\n>>>>>> compat symbols continue to work correctly...  (particularly thinking\n>>>>>> of the messy and totally untested old-FILE support).\n>>>>>\n>>>>> That's the exact purpose of compat_symbol_reference.  I think Carlos is\n>>>>> objecting to its use for a *definition*.\n>>>>\n>>>> Well, I used it for the definitions of matherr and _LIB_VERSION in my\n>>>> tests of those compat symbols, because it does exactly what's expected:\n>>>> makes the definitions in the tests refer to the same entity as the compat\n>>>> symbols in the shared libraries, rather than being completely independent\n>>>> entities as they would by default.\n>>> While it does what's expected, the macro API wasn't designed with that in\n>>> mind was it?\n>>\n>> I would have used it in tst-mallocstate for\n>> __malloc_initialize_hook if I had understood what was going on. I\n>> think the usages are so similar that we do not need a new macro.\n> \n> Carlos, do you still have objections to the patch as posted?\n> \n>   \n> \n> I do not think we need another macro with a different name for this.\n\nI would like to see something like this added:\n\ndiff --git a/include/shlib-compat.h b/include/shlib-compat.h\nindex d872afc..ba99f9b 100644\n--- a/include/shlib-compat.h\n+++ b/include/shlib-compat.h\n@@ -78,8 +78,12 @@\n \n #endif\n \n-/* Use compat_symbol_reference for a reference to a specific version\n- of a symbol. Use compat_symbol to define such a symbol. */\n+/* Use compat_symbol_reference for a reference *or* definition of a \n+ specific version of a symbol. Definitions are primarily used to\n+ ensure tests reference the exact compat symbol required, or define an\n+ interposing symbol of the right version e.g. __malloc_initialize_hook\n+ in mcheck-init.c. Use compat_symbol to define such a symbol within\n+ the shared libraries that are built for users. */\n #define compat_symbol_reference(lib, local, symbol, version) \\\n compat_symbol_reference_1 (lib, local, symbol, version)\n #define compat_symbol_reference_1(lib, local, symbol, version) \\\n---\n\nAnd remove the \"this is a hack\" comments from your patch, since now you\nare using the macro API within the usages of the definition.","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-85666-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"WL3bvNTh\"; dkim-atps=neutral","sourceware.org; auth=none"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yC0bx0JRLz9s81\n\tfor ;\n\tThu, 12 Oct 2017 04:00:44 +1100 (AEDT)","(qmail 8902 invoked by alias); 11 Oct 2017 17:00:39 -0000","(qmail 7049 invoked by uid 89); 11 Oct 2017 17:00:35 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:cc:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=QOtOgNIAjxGKBp0Z\n\t4x/foWH4AWCys8KVAlJ7oqtnecmVvIO20PUQxkJoEPVuZbbkYF6pLC1rs4QOcl7r\n\tEU6oAdy5O4OgtHKCDJkKqK4EKAI23pwcNW9i68RvyVsbAoKf5+HwmpTRHEdRoPH5\n\tRvdgWVS/bOlg6b1ZC+VgTR0wFBI=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:cc:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; s=default; bh=Sr/p98HP/4HyPwwrFYH3Bc\n\tPLCv8=; b=WL3bvNThAvVgvMPZ2nHIWdhxYLr4VdcOUMK3iLPx8W1hyjjSkJehft\n\tbdsIFyCSlxzvQ/8Hr6KdPBl4QVe5EwOfD1LQCEEp2cmXoa846bBcxHPOXluxfHyv\n\t1q8bVMyPjz3WHK3IZlCf0gQVY6493BDAOostwas4j9tAEEu9FgxZI=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-25.4 required=5.0 tests=BAYES_00, GIT_PATCH_0,\n\tGIT_PATCH_1, GIT_PATCH_2, GIT_PATCH_3,\n\tKAM_LAZY_DOMAIN_SECURITY, RCVD_IN_DNSWL_NONE,\n\tRCVD_IN_SORBS_SPAM autolearn=ham version=3.3.2 spammy=","X-HELO":"mail-qt0-f173.google.com","X-Google-DKIM-Signature":"v=1; a=rsa-sha256; c=relaxed/relaxed;\n\td=1e100.net; s=20161025;\n\th=x-gm-message-state:subject:to:cc:references:from:organization\n\t:message-id:date:user-agent:mime-version:in-reply-to\n\t:content-language:content-transfer-encoding;\n\tbh=ZA8w3pNmsPcSYivwvNmyr9A2idmVQ9/cptX65h6sw2w=;\n\tb=n06MAk46o26YhAP5+ImDyPgI3BkPxPc5s6n/DUeHedyc5o5VDG71xw38fXLb11a9V7\n\tHEt6bfB99MUBKTwvBbFFltqePCL1foK/iYZThyxRbVEDSILNiowixIHVZN9Kf0odSftO\n\t1aCHfhz59HZYsTQ7+YIPriYmr3UsZ8tIFUX1eShLMSpLBsJ1idyiSQZCRjlbICls3gLo\n\tlc11ZQKeHDWOnXTb/URYHKEM+i73TkUQy114ThhSxWfkbH7f1oKR3wEQp+tyvRsOD5dV\n\tXlm5Rj4InoY4Bt2RZJKhyuDvMOX2qnnSwL+O4e/bOES/tSZipJzz5RnViaxJ8DRmvht3\n\tujtA==","X-Gm-Message-State":"AMCzsaV/fAJNdyNPlE6hzqET/hJ6fQWQDSceOvyVgNyGfVFc/Y8pJOhu\n\tRWmtW7n2AQhobZnquezH57JTEuDD61k=","X-Google-Smtp-Source":"AOwi7QC8vhhCeCOU1QEGNa94WRjyXxJcUxOmE4/h/0udBOUxYOOWc9pSHUBpqm6GEurLwwKIqGV4fw==","X-Received":"by 10.200.27.171 with SMTP id z40mr467560qtj.151.1507741232004; \n\tWed, 11 Oct 2017 10:00:32 -0700 (PDT)","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","To":"Florian Weimer ,\n\tJoseph Myers ","Cc":"Zack Weinberg ,\n\tGNU C Library ","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>\n\t<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>\n\t\n\t<2ad943f4-1b66-5df7-b348-ab0652716470@redhat.com>\n\t\n\t<61a94166-1768-4014-27f7-34a9a6ff24b8@redhat.com>\n\t\n\t\n\t<5bf3d95b-cdfb-c55f-478f-be61baa846d3@redhat.com>\n\t<834aa29d-d8ad-b516-ab12-98b99ad4faf5@redhat.com>","From":"Carlos O'Donell ","Message-ID":"<60d34595-b3a7-c170-d325-59b1234e3479@redhat.com>","Date":"Wed, 11 Oct 2017 10:00:27 -0700","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.3.0","MIME-Version":"1.0","In-Reply-To":"<834aa29d-d8ad-b516-ab12-98b99ad4faf5@redhat.com>","Content-Type":"text/plain; charset=utf-8","Content-Transfer-Encoding":"8bit"}},{"id":1787649,"web_url":"http://patchwork.ozlabs.org/comment/1787649/","msgid":"","list_archive_url":null,"date":"2017-10-16T18:45:11","subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","submitter":{"id":14312,"url":"http://patchwork.ozlabs.org/api/people/14312/","name":"Florian Weimer","email":"fweimer@redhat.com"},"content":"On 10/11/2017 07:00 PM, Carlos O'Donell wrote:\n> On 10/05/2017 03:45 AM, Florian Weimer wrote:\n>> On 09/18/2017 04:24 PM, Florian Weimer wrote:\n>>> On 08/31/2017 06:08 PM, Carlos O'Donell wrote:\n>>>> On 08/31/2017 11:04 AM, Joseph Myers wrote:\n>>>>> On Thu, 31 Aug 2017, Florian Weimer wrote:\n>>>>>\n>>>>>> On 08/31/2017 05:42 PM, Zack Weinberg wrote:\n>>>>>>> On Thu, Aug 31, 2017 at 11:37 AM, Carlos O'Donell wrote:\n>>>>>>>> I would like to see a new macro that does what it says, rather than use the\n>>>>>>>> existing macro in the wrong way. Even if the new macro is just a copy.\n>>>>>>>>\n>>>>>>>> This looks like a real problem for glibc, particularly if we need to continue\n>>>>>>>> to use, at least internally, certain old versions of symbols. So having a\n>>>>>>>> new macro for this is fine.\n>>>>>>>\n>>>>>>> I see immediate uses for this macro in the test suite, verifying that\n>>>>>>> compat symbols continue to work correctly...  (particularly thinking\n>>>>>>> of the messy and totally untested old-FILE support).\n>>>>>>\n>>>>>> That's the exact purpose of compat_symbol_reference.  I think Carlos is\n>>>>>> objecting to its use for a *definition*.\n>>>>>\n>>>>> Well, I used it for the definitions of matherr and _LIB_VERSION in my\n>>>>> tests of those compat symbols, because it does exactly what's expected:\n>>>>> makes the definitions in the tests refer to the same entity as the compat\n>>>>> symbols in the shared libraries, rather than being completely independent\n>>>>> entities as they would by default.\n>>>> While it does what's expected, the macro API wasn't designed with that in\n>>>> mind was it?\n>>>\n>>> I would have used it in tst-mallocstate for\n>>> __malloc_initialize_hook if I had understood what was going on. I\n>>> think the usages are so similar that we do not need a new macro.\n>>\n>> Carlos, do you still have objections to the patch as posted?\n>>\n>>   \n>>\n>> I do not think we need another macro with a different name for this.\n> \n> I would like to see something like this added:\n> \n> diff --git a/include/shlib-compat.h b/include/shlib-compat.h\n> index d872afc..ba99f9b 100644\n> --- a/include/shlib-compat.h\n> +++ b/include/shlib-compat.h\n> @@ -78,8 +78,12 @@\n> \n> #endif\n> \n> -/* Use compat_symbol_reference for a reference to a specific version\n> - of a symbol. Use compat_symbol to define such a symbol. */\n> +/* Use compat_symbol_reference for a reference *or* definition of a\n> + specific version of a symbol. Definitions are primarily used to\n> + ensure tests reference the exact compat symbol required, or define an\n> + interposing symbol of the right version e.g. __malloc_initialize_hook\n> + in mcheck-init.c. Use compat_symbol to define such a symbol within\n> + the shared libraries that are built for users. */\n> #define compat_symbol_reference(lib, local, symbol, version) \\\n> compat_symbol_reference_1 (lib, local, symbol, version)\n> #define compat_symbol_reference_1(lib, local, symbol, version) \\\n> ---\n> \n> And remove the \"this is a hack\" comments from your patch, since now you\n> are using the macro API within the usages of the definition.\n\nFine, I'm going to push this change (with trailing whitespace removed).\n\nThanks,\nFlorian","headers":{"Return-Path":"","X-Original-To":"incoming@patchwork.ozlabs.org","Delivered-To":["patchwork-incoming@bilbo.ozlabs.org","mailing list libc-alpha@sourceware.org"],"Authentication-Results":["ozlabs.org;\n\tspf=pass (mailfrom) smtp.mailfrom=sourceware.org\n\t(client-ip=209.132.180.131; helo=sourceware.org;\n\tenvelope-from=libc-alpha-return-85899-incoming=patchwork.ozlabs.org@sourceware.org;\n\treceiver=)","ozlabs.org; dkim=pass (1024-bit key;\n\tsecure) header.d=sourceware.org header.i=@sourceware.org\n\theader.b=\"u0dIgkBC\"; dkim-atps=neutral","sourceware.org; auth=none","ext-mx04.extmail.prod.ext.phx2.redhat.com;\n\tdmarc=none (p=none dis=none) header.from=redhat.com","ext-mx04.extmail.prod.ext.phx2.redhat.com;\n\tspf=fail smtp.mailfrom=fweimer@redhat.com"],"Received":["from sourceware.org (server1.sourceware.org [209.132.180.131])\n\t(using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256\n\tbits)) (No client certificate requested)\n\tby ozlabs.org (Postfix) with ESMTPS id 3yG6hp6T03z9s8J\n\tfor ;\n\tTue, 17 Oct 2017 05:45:46 +1100 (AEDT)","(qmail 128925 invoked by alias); 16 Oct 2017 18:45:39 -0000","(qmail 128827 invoked by uid 89); 16 Oct 2017 18:45:24 -0000"],"DomainKey-Signature":"a=rsa-sha1; c=nofws; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:cc:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; q=dns; s=default; b=GhZ/UeWhM3UTxFJJ\n\tCnQcxcPD7mBgrWsFJbtXYyoD3qPKVMnXZ4Z0T+whZZDrcPkuqBIuupcF4mCu30QQ\n\tOO106Amr2IJhooJPS9Cvfz8U4WMIkjM3hVxE2MmMlc+MsCsl5rCMngFBWYGRGJ/9\n\trblzT4Ot2/fj6p/Hmoe2ln8wP78=","DKIM-Signature":"v=1; a=rsa-sha1; c=relaxed; d=sourceware.org; h=list-id\n\t:list-unsubscribe:list-subscribe:list-archive:list-post\n\t:list-help:sender:subject:to:cc:references:from:message-id:date\n\t:mime-version:in-reply-to:content-type\n\t:content-transfer-encoding; s=default; bh=+cO3Ti1xIYCqGxH1QWHQ4C\n\td2g1c=; b=u0dIgkBClm9ZXYF8s4TRPP5gUiK+BqCM6i8Ml+68L0mMx9D1/K0XB0\n\tqyk9ldtruW874KXrpNlTW2JvzgQVPfZTmTX3ZCf3Nz0n7Xo1F+09cBAdKmyY1PW8\n\tttGFgwAVoeQ/RiuFHo8oDTsGlDA4J72dxfnxpJufp/XumH6+59noI=","Mailing-List":"contact libc-alpha-help@sourceware.org; run by ezmlm","Precedence":"bulk","List-Id":"","List-Unsubscribe":"","List-Subscribe":"","List-Archive":"","List-Post":"","List-Help":",\n\t","Sender":"libc-alpha-owner@sourceware.org","X-Virus-Found":"No","X-Spam-SWARE-Status":"No, score=-26.9 required=5.0 tests=BAYES_00, GIT_PATCH_0,\n\tGIT_PATCH_1, GIT_PATCH_2, GIT_PATCH_3, RP_MATCHES_RCVD,\n\tSPF_HELO_PASS autolearn=ham version=3.3.2 spammy=","X-HELO":"mx1.redhat.com","DMARC-Filter":"OpenDMARC Filter v1.3.2 mx1.redhat.com 500A2AD624","Subject":"Re: [PATCH] malloc: Use compat_symbol_reference in libmcheck [BZ\n\t#22050]","To":"Carlos O'Donell ,\n\tJoseph Myers ","Cc":"Zack Weinberg ,\n\tGNU C Library ","References":"<20170831095213.42A4F43994318@oldenburg.str.redhat.com>\n\t<2269f61e-d718-b90e-6d08-4e325dad4ac9@redhat.com>\n\t\n\t<2ad943f4-1b66-5df7-b348-ab0652716470@redhat.com>\n\t\n\t<61a94166-1768-4014-27f7-34a9a6ff24b8@redhat.com>\n\t\n\t\n\t<5bf3d95b-cdfb-c55f-478f-be61baa846d3@redhat.com>\n\t<834aa29d-d8ad-b516-ab12-98b99ad4faf5@redhat.com>\n\t<60d34595-b3a7-c170-d325-59b1234e3479@redhat.com>","From":"Florian Weimer ","Message-ID":"","Date":"Mon, 16 Oct 2017 20:45:11 +0200","User-Agent":"Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101\n\tThunderbird/52.3.0","MIME-Version":"1.0","In-Reply-To":"<60d34595-b3a7-c170-d325-59b1234e3479@redhat.com>","Content-Type":"text/plain; charset=utf-8; format=flowed","Content-Transfer-Encoding":"8bit"}}]