From: Adhemerval Zanella <adhemerval.zanella@linaro.org>
To: Florian Weimer <fweimer@redhat.com>
Cc: libc-alpha@sourceware.org, John Mellor-Crummey <johnmc@rice.edu>,
Ben Woodard <woodard@redhat.com>
Subject: Re: [PATCH v5 05/22] elf: Fix initial-exec TLS access on audit modules (BZ #28096)
Date: Thu, 11 Nov 2021 15:54:34 -0300 [thread overview]
Message-ID: <bee305bb-3a12-b937-e5f3-4ee557050de8@linaro.org> (raw)
In-Reply-To: <87v9105el8.fsf@oldenburg.str.redhat.com>
On 10/11/2021 10:23, Florian Weimer wrote:
> * Adhemerval Zanella:
>
>> diff --git a/elf/Makefile b/elf/Makefile
>> index 4758cb23c4..861e83236e 100644
>> --- a/elf/Makefile
>> +++ b/elf/Makefile
>
>> +$(objpfx)tst-audit21.out: $(objpfx)tst-auditmod21.so
>> +tst-audit21-ENV = LD_AUDIT=$(objpfx)tst-auditmod21.so
>
> This is needed to fix a Hurd build failure:
>
> $(objpfx)tst-audit21: $(shared-thread-library)
Ack.
>
>> diff --git a/elf/dl-object.c b/elf/dl-object.c
>> index 1875599eb2..eb2158a84b 100644
>> --- a/elf/dl-object.c
>> +++ b/elf/dl-object.c
>> @@ -175,6 +175,9 @@ _dl_new_object (char *realname, const char *libname, int type,
>>
>> new->l_local_scope[0] = &new->l_searchlist;
>>
>> + if (mode & __RTLD_AUDIT)
>> + new->l_dont_set_tls_static = 1;
>> +
>> /* Determine the origin. If allocating the link map for the main
>> executable, the realname is not known and "". In this case, the
>> origin needs to be determined by other means. However, in case
>
>> diff --git a/elf/rtld.c b/elf/rtld.c
>> index 8953347b00..db1817655f 100644
>> --- a/elf/rtld.c
>> +++ b/elf/rtld.c
>> @@ -1055,6 +1055,8 @@ ERROR: audit interface '%s' requires version %d (maximum supported version %d);
>>
>> /* Mark the DSO as being used for auditing. */
>> dlmargs.map->l_auditing = 1;
>> + /* Mark the DSO to not clear the TLS bss in tls initialization. */
>> + dlmargs.map->l_dont_set_tls_static = 1;
>> }
>>
>> /* Notify the the audit modules that the object MAP has already been
>
> I'm not sure if this actually works. As far as I understand it,
> everything in an audit namespace needs this special treatment, and not
> just the audit module itself.
Yes, and since each audit modules is loaded with LM_ID_NEWLM it means
all its dependencies. Adding some instrumentation, it works on the audit
modules does have some dependencies:
$ ldd /home/azanella/Projects/glibc/build/x86_64-linux-gnu/elf/tst-auditmod21a.so
linux-vdso.so.1 (0x00007fff4a172000)
/home/azanella/Projects/glibc/build/x86_64-linux-gnu/elf/tst-auditmod21b.so (0x00007fcd11fa5000)
libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007fcd11d61000)
/lib64/ld-linux-x86-64.so.2 (0x00007fcd11fb1000)
$ LD_AUDIT=elf/tst-auditmod21a.so elf/ld.so --library-path . elf/tst-audit21
2281374: [../elf/dl-tls.c:603] map->l_name= l_dont_set_tls_static=0
2281374: [../elf/dl-tls.c:603] map->l_name=elf/tst-auditmod21a.so l_dont_set_tls_static=1
2281374: [../elf/dl-tls.c:603] map->l_name=/home/azanella/Projects/glibc/build/x86_64-linux-gnu/elf/tst-auditmod21b.so l_dont_set_tls_static=1
2281374: [../elf/dl-tls.c:603] map->l_name=./libc.so.6 l_dont_set_tls_static=1
2281374: [../elf/dl-tls.c:603] map->l_name=/home/azanella/Projects/glibc/build/x86_64-linux-gnu/elf/tst-auditmod21c.so l_dont_set_tls_static=0
2281374: [../elf/dl-tls.c:603] map->l_name=./libc.so.6 l_dont_set_tls_static=0
2281374: [../elf/dl-tls.c:603] map->l_name= l_dont_set_tls_static=0
2281374: [../elf/dl-tls.c:603] map->l_name=elf/tst-auditmod21a.so l_dont_set_tls_static=0
2281374: [../elf/dl-tls.c:603] map->l_name=/home/azanella/Projects/glibc/build/x86_64-linux-gnu/elf/tst-auditmod21b.so l_dont_set_tls_static=0
2281374: [../elf/dl-tls.c:603] map->l_name=./libc.so.6 l_dont_set_tls_static=0
2281374: [../elf/dl-tls.c:603] map->l_name=/home/azanella/Projects/glibc/build/x86_64-linux-gnu/elf/tst-auditmod21c.so l_dont_set_tls_static=0
2281374: [../elf/dl-tls.c:603] map->l_name=./libc.so.6 l_dont_set_tls_static=0
The l_dont_set_tls_static should be set for the audit module and its
dependencies.
>
> I think we shuld add a parameter to _dl_allocate_tls_init and that
> indicates the initialization should only be applied to objects in the
> base namespace. This way, initialization is also skipped for dlopen'ed
> modules in an audit namespace.
I am not sure if this would be best solution, I will check whether
dlopen does work with this patch.
next prev parent reply other threads:[~2021-11-11 18:54 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-11-09 18:33 [PATCH v5 00/22] Multiple rtld-audit fixes Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 01/22] elf: Avoid unnecessary slowdown from profiling with audit (BZ#15533) Adhemerval Zanella
2021-11-10 12:11 ` Florian Weimer
2021-11-10 19:53 ` Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 02/22] elf: Add audit tests for modules with TLSDESC Adhemerval Zanella
2021-11-10 13:55 ` Florian Weimer
2021-11-11 19:18 ` Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 03/22] elf: Do not fail for failed dlopem on audit modules (BZ #28061) Adhemerval Zanella
2021-11-09 18:51 ` H.J. Lu
2021-11-11 17:24 ` Adhemerval Zanella
2021-11-10 14:00 ` Florian Weimer
2021-11-11 17:29 ` Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 04/22] elf: Suppress audit calls when a (new) namespace is empty (BZ #28062) Adhemerval Zanella
2021-11-10 14:15 ` Florian Weimer
2021-11-11 11:51 ` Adhemerval Zanella
2021-11-11 12:02 ` Florian Weimer
2021-11-11 12:25 ` Adhemerval Zanella
2021-11-11 12:33 ` Florian Weimer
2021-11-11 13:02 ` Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 05/22] elf: Fix initial-exec TLS access on audit modules (BZ #28096) Adhemerval Zanella
2021-11-10 13:23 ` Florian Weimer
2021-11-11 18:54 ` Adhemerval Zanella [this message]
2021-11-09 18:33 ` [PATCH v5 06/22] elf: Add _dl_audit_objopen Adhemerval Zanella
2021-11-10 16:51 ` Florian Weimer
2021-11-09 18:33 ` [PATCH v5 07/22] elf: Add _dl_audit_activity_map and _dl_audit_activity_nsid Adhemerval Zanella
2021-11-10 16:59 ` Florian Weimer
2021-11-09 18:33 ` [PATCH v5 08/22] elf: Add _dl_audit_objsearch Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 09/22] elf: Add _dl_audit_objclose Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 10/22] elf: Add _dl_audit_symbind_alt and _dl_audit_symbind Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 11/22] elf: Add _dl_audit_preinit Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 12/22] elf: Add _dl_audit_pltenter Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 13/22] elf: Add _dl_audit_pltexit Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 14/22] elf: Issue audit la_objopen() for vDSO Adhemerval Zanella
2021-11-11 17:50 ` Florian Weimer
2021-11-11 20:16 ` Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 15/22] elf: Run constructors if executable has a soname of a dependency Adhemerval Zanella
2021-11-11 12:30 ` Florian Weimer
2021-11-12 19:02 ` Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 16/22] elf: Add main application on main_map l_name Adhemerval Zanella
2021-11-11 12:39 ` Florian Weimer
2021-11-12 19:30 ` Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 17/22] elf: Add la_activity during application exit Adhemerval Zanella
2021-11-11 12:50 ` Florian Weimer
2021-11-12 19:32 ` Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 18/22] elf: Issue la_symbind() for bind-now (BZ #23734) Adhemerval Zanella
2021-11-11 17:39 ` Florian Weimer
2021-11-15 14:20 ` Adhemerval Zanella
2021-11-15 14:23 ` Florian Weimer
2021-11-15 15:54 ` Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 19/22] elf: Add LA_SYMB_BINDNOW Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 20/22] elf: Move LAV_CURRENT to link_lavcurrent.h Adhemerval Zanella
2021-11-11 17:42 ` Florian Weimer
2021-11-15 14:21 ` Adhemerval Zanella
2021-11-09 18:33 ` [PATCH v5 21/22] elf: Fix runtime linker auditing on aarch64 (BZ #26643) Adhemerval Zanella
2021-11-09 18:33 ` [RFC v5 22/22] elf: Add SVE support for aarch64 rtld-audit Adhemerval Zanella
2021-11-10 13:52 ` Florian Weimer
2021-11-15 17:04 ` Adhemerval Zanella
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=bee305bb-3a12-b937-e5f3-4ee557050de8@linaro.org \
--to=adhemerval.zanella@linaro.org \
--cc=fweimer@redhat.com \
--cc=johnmc@rice.edu \
--cc=libc-alpha@sourceware.org \
--cc=woodard@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).