| Message ID | 87ft0wybrj.fsf@oldenburg.str.redhat.com |
|---|---|
| State | Superseded |
| Headers |
Return-Path: <libc-alpha-bounces@sourceware.org> X-Original-To: patchwork@sourceware.org Delivered-To: patchwork@sourceware.org Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id 5F5A93858022; Mon, 15 Mar 2021 10:03:32 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 5F5A93858022 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=sourceware.org; s=default; t=1615802612; bh=hByTsQUMhkEAQmHvjWH9Po1JwNjz4S3s8GOxYq2hE9g=; h=To:Subject:Date:List-Id:List-Unsubscribe:List-Archive:List-Post: List-Help:List-Subscribe:From:Reply-To:From; b=kJmDbZVJCTEk7NF3LZLE5iPoJK2Laj4IOiASOFtBi0Jg1NQLCxVneVGIahWXwNNta o0QzuqQEyL4kg1EnLI/7C6lePCrXbrLceorK5JQBfD4jl2o89Ffu752rji0H+CzzG6 CQLxawXn5Yzbct4FuU1wt52OEG1bO6d/CMwid5HM= X-Original-To: libc-alpha@sourceware.org Delivered-To: libc-alpha@sourceware.org Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [216.205.24.124]) by sourceware.org (Postfix) with ESMTP id 8DEBC3858C27 for <libc-alpha@sourceware.org>; Mon, 15 Mar 2021 10:03:29 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.3.2 sourceware.org 8DEBC3858C27 Received: from mimecast-mx01.redhat.com (mimecast-mx01.redhat.com [209.132.183.4]) (Using TLS) by relay.mimecast.com with ESMTP id us-mta-291-CJY7kGJUMTy3tbfcTFF1ZQ-1; Mon, 15 Mar 2021 06:03:24 -0400 X-MC-Unique: CJY7kGJUMTy3tbfcTFF1ZQ-1 Received: from smtp.corp.redhat.com (int-mx02.intmail.prod.int.phx2.redhat.com [10.5.11.12]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx01.redhat.com (Postfix) with ESMTPS id 8EE26101F001 for <libc-alpha@sourceware.org>; Mon, 15 Mar 2021 10:03:23 +0000 (UTC) Received: from oldenburg.str.redhat.com (ovpn-112-163.ams2.redhat.com [10.36.112.163]) by smtp.corp.redhat.com (Postfix) with ESMTPS id EA6BE60C5F for <libc-alpha@sourceware.org>; Mon, 15 Mar 2021 10:03:22 +0000 (UTC) To: libc-alpha@sourceware.org Subject: [PATCH] elf: ld.so --help calls _dl_init_paths without a main map [BZ #27577] Date: Mon, 15 Mar 2021 11:03:28 +0100 Message-ID: <87ft0wybrj.fsf@oldenburg.str.redhat.com> User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/27.1 (gnu/linux) MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.79 on 10.5.11.12 X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com Content-Type: text/plain X-Spam-Status: No, score=-12.5 required=5.0 tests=BAYES_00, DKIMWL_WL_HIGH, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, GIT_PATCH_0, RCVD_IN_DNSWL_LOW, RCVD_IN_MSPIKE_H3, RCVD_IN_MSPIKE_WL, SPF_HELO_NONE, SPF_PASS, TXREP autolearn=ham autolearn_force=no version=3.4.2 X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on server2.sourceware.org X-BeenThere: libc-alpha@sourceware.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Libc-alpha mailing list <libc-alpha.sourceware.org> List-Unsubscribe: <https://sourceware.org/mailman/options/libc-alpha>, <mailto:libc-alpha-request@sourceware.org?subject=unsubscribe> List-Archive: <https://sourceware.org/pipermail/libc-alpha/> List-Post: <mailto:libc-alpha@sourceware.org> List-Help: <mailto:libc-alpha-request@sourceware.org?subject=help> List-Subscribe: <https://sourceware.org/mailman/listinfo/libc-alpha>, <mailto:libc-alpha-request@sourceware.org?subject=subscribe> From: Florian Weimer via Libc-alpha <libc-alpha@sourceware.org> Reply-To: Florian Weimer <fweimer@redhat.com> Errors-To: libc-alpha-bounces@sourceware.org Sender: "Libc-alpha" <libc-alpha-bounces@sourceware.org> |
| Series |
elf: ld.so --help calls _dl_init_paths without a main map [BZ #27577]
|
|
Commit Message
Florian Weimer
March 15, 2021, 10:03 a.m. UTC
In this case, use the link map of the dynamic loader itself as
a replacement. This is more than just a hack: if we ever support
DT_RUNPATH/DT_RPATH for the dynamic loader, reporting it for
ld.so --help (without further command line arguments) would be the
right thing to do.
Fixes commit 332421312576bd7095e70589154af99b124dd2d1 ("elf: Always
set l in _dl_init_paths (bug 23462)").
Tested on i686-linux-gnu and x86_64-linux-gnu.
---
elf/Makefile | 10 +++++++++-
elf/dl-load.c | 8 +++++++-
2 files changed, 16 insertions(+), 2 deletions(-)
Comments
On Mon, Mar 15, 2021 at 3:03 AM Florian Weimer via Libc-alpha <libc-alpha@sourceware.org> wrote: > > In this case, use the link map of the dynamic loader itself as > a replacement. This is more than just a hack: if we ever support > DT_RUNPATH/DT_RPATH for the dynamic loader, reporting it for > ld.so --help (without further command line arguments) would be the > right thing to do. > > Fixes commit 332421312576bd7095e70589154af99b124dd2d1 ("elf: Always > set l in _dl_init_paths (bug 23462)"). > > Tested on i686-linux-gnu and x86_64-linux-gnu. > > --- > elf/Makefile | 10 +++++++++- > elf/dl-load.c | 8 +++++++- > 2 files changed, 16 insertions(+), 2 deletions(-) > > diff --git a/elf/Makefile b/elf/Makefile > index 4c9e63dac9..ebb331df36 100644 > --- a/elf/Makefile > +++ b/elf/Makefile > @@ -245,7 +245,8 @@ tests += $(tests-execstack-$(have-z-execstack)) > ifeq ($(run-built-tests),yes) > tests-special += $(objpfx)tst-leaks1-mem.out \ > $(objpfx)tst-leaks1-static-mem.out $(objpfx)noload-mem.out \ > - $(objpfx)tst-ldconfig-X.out > + $(objpfx)tst-ldconfig-X.out $(objpfx)tst-rtld-help.out \ > + $(objpfx)tst-rtld-help-2.out > endif > tlsmod17a-suffixes = 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 > tlsmod18a-suffixes = 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 > @@ -1907,3 +1908,10 @@ $(objpfx)list-tunables.out: tst-rtld-list-tunables.sh $(objpfx)ld.so > $(evaluate-test) > > tst-dst-static-ENV = LD_LIBRARY_PATH='$$ORIGIN' > + > +$(objpfx)tst-rtld-help.out: $(objpfx)ld.so > + $(test-wrapper) $(rtld-prefix) --help > $@; \ > + $(evaluate-test) > +$(objpfx)tst-rtld-help-2.out: $(objpfx)tst-rtld-help.out > + grep -q 'Legacy HWCAP subdirectories under library search path directories' $<; \ > + $(evaluate-test) Can we avoid tst-rtld-help-2.out by grep $@ directly? > diff --git a/elf/dl-load.c b/elf/dl-load.c > index 376a2e64d6..2f760503c5 100644 > --- a/elf/dl-load.c > +++ b/elf/dl-load.c > @@ -758,8 +758,14 @@ _dl_init_paths (const char *llp, const char *source, > max_dirnamelen = SYSTEM_DIRS_MAX_LEN; > *aelem = NULL; > > - /* This points to the map of the main object. */ > + /* This points to the map of the main object. If there is no main > + object (e.g., under --help, use the dynamic loader itself as a > + stand-in. */ > l = GL(dl_ns)[LM_ID_BASE]._ns_loaded; > +#ifdef SHARED > + if (l == NULL) > + l = &GL (dl_rtld_map); > +#endif > assert (l->l_type != lt_loaded); > > if (l->l_info[DT_RUNPATH]) >
* H. J. Lu: >> +$(objpfx)tst-rtld-help.out: $(objpfx)ld.so >> + $(test-wrapper) $(rtld-prefix) --help > $@; \ >> + $(evaluate-test) >> +$(objpfx)tst-rtld-help-2.out: $(objpfx)tst-rtld-help.out >> + grep -q 'Legacy HWCAP subdirectories under library search path directories' $<; \ >> + $(evaluate-test) > > Can we avoid tst-rtld-help-2.out by grep $@ directly? We could use &&, but that would make diagnosing test failures difficult (as in: did --help fail, or grep?). The separate test seemed the easiest way. Thanks, Florian
On Mon, Mar 15, 2021 at 5:14 AM Florian Weimer <fweimer@redhat.com> wrote: > > * H. J. Lu: > > >> +$(objpfx)tst-rtld-help.out: $(objpfx)ld.so > >> + $(test-wrapper) $(rtld-prefix) --help > $@; \ > >> + $(evaluate-test) > >> +$(objpfx)tst-rtld-help-2.out: $(objpfx)tst-rtld-help.out > >> + grep -q 'Legacy HWCAP subdirectories under library search path directories' $<; \ > >> + $(evaluate-test) > > > > Can we avoid tst-rtld-help-2.out by grep $@ directly? > > We could use &&, but that would make diagnosing test failures difficult > (as in: did --help fail, or grep?). The separate test seemed the > easiest way. > Running "ld.so --help" itself is very straight forward. When there is foo.out, I usually search for foo.
* H. J. Lu: > On Mon, Mar 15, 2021 at 5:14 AM Florian Weimer <fweimer@redhat.com> wrote: >> >> * H. J. Lu: >> >> >> +$(objpfx)tst-rtld-help.out: $(objpfx)ld.so >> >> + $(test-wrapper) $(rtld-prefix) --help > $@; \ >> >> + $(evaluate-test) >> >> +$(objpfx)tst-rtld-help-2.out: $(objpfx)tst-rtld-help.out >> >> + grep -q 'Legacy HWCAP subdirectories under library search path directories' $<; \ >> >> + $(evaluate-test) >> > >> > Can we avoid tst-rtld-help-2.out by grep $@ directly? >> >> We could use &&, but that would make diagnosing test failures difficult >> (as in: did --help fail, or grep?). The separate test seemed the >> easiest way. > > Running "ld.so --help" itself is very straight forward. When there is > foo.out, I usually search for foo. Sorry, I don't understand what your comment means in this context. Would you please elaborate? Thanks, Florian
On Mon, Mar 15, 2021 at 5:28 AM Florian Weimer <fweimer@redhat.com> wrote: > > * H. J. Lu: > > > On Mon, Mar 15, 2021 at 5:14 AM Florian Weimer <fweimer@redhat.com> wrote: > >> > >> * H. J. Lu: > >> > >> >> +$(objpfx)tst-rtld-help.out: $(objpfx)ld.so > >> >> + $(test-wrapper) $(rtld-prefix) --help > $@; \ > >> >> + $(evaluate-test) > >> >> +$(objpfx)tst-rtld-help-2.out: $(objpfx)tst-rtld-help.out > >> >> + grep -q 'Legacy HWCAP subdirectories under library search path directories' $<; \ > >> >> + $(evaluate-test) > >> > > >> > Can we avoid tst-rtld-help-2.out by grep $@ directly? > >> > >> We could use &&, but that would make diagnosing test failures difficult > >> (as in: did --help fail, or grep?). The separate test seemed the > >> easiest way. > > > > Running "ld.so --help" itself is very straight forward. When there is > > foo.out, I usually search for foo. > > Sorry, I don't understand what your comment means in this context. > Would you please elaborate? A single tst-rtld-help.out doesn't make diagnosing ld.so --help failures difficult. Multiple output files is confusing.
On Mär 15 2021, Florian Weimer via Libc-alpha wrote: > +$(objpfx)tst-rtld-help-2.out: $(objpfx)tst-rtld-help.out > + grep -q 'Legacy HWCAP subdirectories under library search path directories' $<; \ > + $(evaluate-test) Shouldn't that create the target? Andreas.
* H. J. Lu: > On Mon, Mar 15, 2021 at 5:28 AM Florian Weimer <fweimer@redhat.com> wrote: >> >> * H. J. Lu: >> >> > On Mon, Mar 15, 2021 at 5:14 AM Florian Weimer <fweimer@redhat.com> wrote: >> >> >> >> * H. J. Lu: >> >> >> >> >> +$(objpfx)tst-rtld-help.out: $(objpfx)ld.so >> >> >> + $(test-wrapper) $(rtld-prefix) --help > $@; \ >> >> >> + $(evaluate-test) >> >> >> +$(objpfx)tst-rtld-help-2.out: $(objpfx)tst-rtld-help.out >> >> >> + grep -q 'Legacy HWCAP subdirectories under library search path directories' $<; \ >> >> >> + $(evaluate-test) >> >> > >> >> > Can we avoid tst-rtld-help-2.out by grep $@ directly? >> >> >> >> We could use &&, but that would make diagnosing test failures difficult >> >> (as in: did --help fail, or grep?). The separate test seemed the >> >> easiest way. >> > >> > Running "ld.so --help" itself is very straight forward. When there is >> > foo.out, I usually search for foo. >> >> Sorry, I don't understand what your comment means in this context. >> Would you please elaborate? > > A single tst-rtld-help.out doesn't make diagnosing ld.so --help > failures difficult. > Multiple output files is confusing. All mtrace tests do this. However, I will submit a different patch with just one target. Thanks, Florian
* Andreas Schwab: > On Mär 15 2021, Florian Weimer via Libc-alpha wrote: > >> +$(objpfx)tst-rtld-help-2.out: $(objpfx)tst-rtld-help.out >> + grep -q 'Legacy HWCAP subdirectories under library search path directories' $<; \ >> + $(evaluate-test) > > Shouldn't that create the target? H.J. wants me to remove the separate target. It will address this issue as well. Thanks, Florian
diff --git a/elf/Makefile b/elf/Makefile index 4c9e63dac9..ebb331df36 100644 --- a/elf/Makefile +++ b/elf/Makefile @@ -245,7 +245,8 @@ tests += $(tests-execstack-$(have-z-execstack)) ifeq ($(run-built-tests),yes) tests-special += $(objpfx)tst-leaks1-mem.out \ $(objpfx)tst-leaks1-static-mem.out $(objpfx)noload-mem.out \ - $(objpfx)tst-ldconfig-X.out + $(objpfx)tst-ldconfig-X.out $(objpfx)tst-rtld-help.out \ + $(objpfx)tst-rtld-help-2.out endif tlsmod17a-suffixes = 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 tlsmod18a-suffixes = 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 @@ -1907,3 +1908,10 @@ $(objpfx)list-tunables.out: tst-rtld-list-tunables.sh $(objpfx)ld.so $(evaluate-test) tst-dst-static-ENV = LD_LIBRARY_PATH='$$ORIGIN' + +$(objpfx)tst-rtld-help.out: $(objpfx)ld.so + $(test-wrapper) $(rtld-prefix) --help > $@; \ + $(evaluate-test) +$(objpfx)tst-rtld-help-2.out: $(objpfx)tst-rtld-help.out + grep -q 'Legacy HWCAP subdirectories under library search path directories' $<; \ + $(evaluate-test) diff --git a/elf/dl-load.c b/elf/dl-load.c index 376a2e64d6..2f760503c5 100644 --- a/elf/dl-load.c +++ b/elf/dl-load.c @@ -758,8 +758,14 @@ _dl_init_paths (const char *llp, const char *source, max_dirnamelen = SYSTEM_DIRS_MAX_LEN; *aelem = NULL; - /* This points to the map of the main object. */ + /* This points to the map of the main object. If there is no main + object (e.g., under --help, use the dynamic loader itself as a + stand-in. */ l = GL(dl_ns)[LM_ID_BASE]._ns_loaded; +#ifdef SHARED + if (l == NULL) + l = &GL (dl_rtld_map); +#endif assert (l->l_type != lt_loaded); if (l->l_info[DT_RUNPATH])