This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: Async-signal-safe access to __thread variables from dlopen()ed libraries?
- From: "Carlos O'Donell" <carlos at redhat dot com>
- To: Ian Lance Taylor <iant at google dot com>
- Cc: Paul Pluzhnikov <ppluzhnikov at google dot com>, Roland McGrath <roland at hack dot frob dot com>, Richard Henderson <rth at twiddle dot net>, GNU C Library <libc-alpha at sourceware dot org>, Andrew Hunter <ahh at google dot com>, Alexandre Oliva <aoliva at redhat dot com>
- Date: Tue, 24 Sep 2013 16:15:11 -0400
- Subject: Re: Async-signal-safe access to __thread variables from dlopen()ed libraries?
- Authentication-results: sourceware.org; auth=none
- References: <20120612193224 dot 8E43619060E at elbrus2 dot mtv dot corp dot google dot com> <4FD8D974 dot 7090903 at twiddle dot net> <20120613182826 dot 0CFAB2C0A3 at topped-with-meat dot com> <CALoOobMtXCw+oe7ZL0=my8YH5st8b1==CasS8i07z6G9DfaX-w at mail dot gmail dot com> <20120613210444 dot 659732C095 at topped-with-meat dot com> <mcr4nqebzok dot fsf at dhcp-172-18-216-180 dot mtv dot corp dot google dot com> <20120614002931 dot ABB762C08B at topped-with-meat dot com> <mcr1uliaeep dot fsf at dhcp-172-18-216-180 dot mtv dot corp dot google dot com> <CALoOobPJ7G7ciRfc2JwzHjsDTg4-_h-SXqeU1zR4WEzoyQhyNg at mail dot gmail dot com> <523BD470 dot 6090203 at redhat dot com> <CAKOQZ8y85QBkd97cEEmP-4OgE2KizCqknrVR_n44pwBGMs5uAw at mail dot gmail dot com> <523C88D1 dot 6090304 at redhat dot com> <CAKOQZ8ze1zKdQRsMsmBCqnJr361Gvv8mYjLjGgzYwWJEKUY+7w at mail dot gmail dot com> <523CB72A dot 7040507 at redhat dot com> <CAKOQZ8yq6gnWig3Wg4YF0qOYTJbevEExG0Sm9K4ofsSO+PWq1A at mail dot gmail dot com>
On 09/20/2013 05:28 PM, Ian Lance Taylor wrote:
> On Fri, Sep 20, 2013 at 1:59 PM, Carlos O'Donell <carlos@redhat.com> wrote:
>>
>> I've started a wiki page and I'll summarize people's comments there:
>> https://sourceware.org/glibc/wiki/TLSandSignals
>
> I tried to edit the page but I evidently don't have permissions (I
> just created an account on the wiki IanLanceTaylor).
>
> I don't think you characterized the problem precisely. The issue is
> not receiving a signal while the thread is initializing TLS. The
> general issue is that calling __tls_get_addr can call malloc, which is
> not async-signal-safe. A specific example would be
>
> * There is an existing thread.
> * The program does a dlopen of a shared library that uses TLS.
> * The shared library installs a signal handler that refers to a TLS variable.
> * The signal handler is called on the existing thread.
> * Because shared library TLS variables are installed lazily, the
> existing thread does not yet have a copy of the TLS variable.
> * Therefore the signal handlers TLS reference calls malloc.
> * If the signal occurred during a call to malloc, we have a deadlock.
Thanks. I still think we need to look at interrupted initialization
as another possible problem.
Cheers,
Carlos.