unofficial mirror of libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: Szabolcs Nagy via Libc-alpha <libc-alpha@sourceware.org>
To: Fangrui Song <maskray@google.com>
Cc: libc-alpha@sourceware.org
Subject: Re: [PATCH] aarch64: Skip traditional GD/LD TLS which are unsupported by Clang and LLD [BZ #28205]
Date: Tue, 10 Aug 2021 12:20:03 +0100	[thread overview]
Message-ID: <20210810112002.GH20410@arm.com> (raw)
In-Reply-To: <20210806224126.643993-1-maskray@google.com>

The 08/06/2021 15:41, Fangrui Song via Libc-alpha wrote:
> TLSDESC is the default on aarch64.  Clang doesn't support
> -mtls-dialect=trad.  Its integrated assembler doesn't support the
> marker.  LLD's doesn't support R_AARCH64_TLSGD_*/R_AARCH64_TLSLD_*
> relocations.  Just skip the tests.
> 
> With https://sourceware.org/pipermail/libc-alpha/2021-August/129904.html
> ("aarch64: Make elf_machine_{load_address, dynamic} robust [BZ #28203]"),
> if we allow LLD in configure.ac,
> `make check` test results of LLD are on par with GNU ld.
> ---
>  config.h.in       |  3 +++
>  configure         | 38 ++++++++++++++++++++++++++++++++++++++
>  configure.ac      | 24 ++++++++++++++++++++++++
>  elf/tst-tls1.c    |  7 +++++--
>  elf/tst-tls2.c    |  6 ++++--
>  elf/tst-tls3.c    |  8 ++++----
>  elf/tst-tlsmod1.c |  6 ++++--
>  elf/tst-tlsmod2.c |  4 +++-
>  elf/tst-tlsmod3.c |  5 ++++-
>  elf/tst-tlsmod4.c |  4 +++-
>  10 files changed, 92 insertions(+), 13 deletions(-)
> 
> diff --git a/config.h.in b/config.h.in
> index 8b45a3a61d..f65605d4bd 100644
> --- a/config.h.in
> +++ b/config.h.in
> @@ -198,6 +198,9 @@
>  /* Define if CC supports attribute retain.  */
>  #undef HAVE_GNU_RETAIN
>  
> +/* Define if CC and LD support -mtls-dialect=trad.  */
> +#undef HAVE_MTLS_DIALECT_TRAD
> +
>  /* Define if the linker defines __ehdr_start.  */
>  #undef HAVE_EHDR_START
>  
> diff --git a/configure.ac b/configure.ac
> index 5632277f9c..c73e271cc8 100644
> --- a/configure.ac
> +++ b/configure.ac
> @@ -1501,6 +1501,30 @@ rm -f conftest*])
>  AC_SUBST(libc_cv_mtls_dialect_gnu2)
>  LIBC_CONFIG_VAR([have-mtls-dialect-gnu2], [$libc_cv_mtls_dialect_gnu2])
>  
> +AC_CACHE_CHECK([for -mtls-dialect=trad], libc_cv_mtls_dialect_trad,
> +[dnl
> +cat > conftest.c <<EOF
> +extern __thread int i;
> +void foo (void)
> +{
> +  i = 10;
> +}
> +EOF
> +if AC_TRY_COMMAND([${CC-cc} $CFLAGS $CPPFLAGS $LDFLAGS
> +			-fPIC -mtls-dialect=trad -shared -o conftest.so conftest.c
> +			1>&AS_MESSAGE_LOG_FD])
> +then
> +  libc_cv_mtls_dialect_trad=yes
> +else
> +  libc_cv_mtls_dialect_trad=no
> +fi
> +rm -f conftest*])
> +if test $libc_cv_mtls_dialect_trad = yes; then
> +  AC_DEFINE(HAVE_MTLS_DIALECT_TRAD)
> +fi
> +AC_SUBST(libc_cv_mtls_dialect_trad)
> +LIBC_CONFIG_VAR([have-mtls-dialect-trad], [$libc_cv_mtls_dialect_trad])

this is a possible way, but i think it's nicer to have

#define HAVE_TRAD_TLS 1

in config.h.in, and then in sysdeps/aarch64/configure.ac

AC_DEFINE(HAVE_TRAD_TLS, 0, [Only TLSDESC is supported])

(haven't tested, but i think it would work)

(-mtls-dialect=trad is aarch64 specific flag, so it's
not nice in the generic configure and the tests don't
actually care about this flag but whether the assembler
and linker would work)



      reply	other threads:[~2021-08-10 11:20 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-06 22:41 [PATCH] aarch64: Skip traditional GD/LD TLS which are unsupported by Clang and LLD [BZ #28205] Fangrui Song via Libc-alpha
2021-08-10 11:20 ` Szabolcs Nagy via Libc-alpha [this message]

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

  List information: https://www.gnu.org/software/libc/involved.html

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210810112002.GH20410@arm.com \
    --to=libc-alpha@sourceware.org \
    --cc=maskray@google.com \
    --cc=szabolcs.nagy@arm.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).