From: "Marc-André Lureau" <marcandre.lureau@redhat.com>
To: "Emilio G. Cota" <cota@braap.org>
Cc: qemu-devel <qemu-devel@nongnu.org>, Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [Qemu-devel] [PATCH] configure: fix sanitizers' test program to mend ASan detection
Date: Wed, 21 Feb 2018 10:12:28 +0100 [thread overview]
Message-ID: <CAMxuvazD3r_-soineNOs1ZEdfUZdrYuCOq5j2xJVb0ZY5k7iaQ@mail.gmail.com> (raw)
In-Reply-To: <1519174995-1039-1-git-send-email-cota@braap.org>
Hi
On Wed, Feb 21, 2018 at 2:03 AM, Emilio G. Cota <cota@braap.org> wrote:
> Commit 218bb57 ("build-sys: check static linking of UBSAN", 2018-02-13)
> adds a small test program to check whether ubsan works even when
> configuring with --static. This added program is used to
> detect all sanitizers, which breaks ASan's detection since the
> compilation fails with -fsanitize=address, at least with gcc 5.4.0
> and 7.2.0:
>
> qemu-conf.c: In function ‘main’:
> qemu-conf.c:3:20: error: integer overflow in expression [-Werror=overflow]
> return INT32_MIN / -1;
> ^
> cc1: all warnings being treated as errors
I sent a patch a few days ago:
"[PATCH 1/6] build-sys: fix -fsanitize=address check"
>
> Fix it by:
>
> - Changing the test program to one that incurs undefined behaviour that isn't
> detected at compile-time, even with -fsanitize=address.
>
> - To be extra safe (since compilers might evolve and eventually figure out
> the UB at compile-time), use this newly-added test only when checking
> for UBSan; use the skeleton otherwise.
>
> Signed-off-by: Emilio G. Cota <cota@braap.org>
> ---
> configure | 20 ++++++++++++--------
> 1 file changed, 12 insertions(+), 8 deletions(-)
>
> diff --git a/configure b/configure
> index 913e148..56f647f 100755
> --- a/configure
> +++ b/configure
> @@ -5306,13 +5306,13 @@ fi
> ##########################################
> # checks for sanitizers
>
> -# we could use a simple skeleton for flags checks, but this also
> -# detect the static linking issue of ubsan, see also:
> +# Use this program to detect the static linking issue of ubsan; see
> # https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84285
> cat > $TMPC << EOF
> -#include <stdint.h>
> -int main(void) {
> - return INT32_MIN / -1;
> +int main(int argc, char **argv) {
> + int k = 0x7fffffff;
> + k += argc;
> + return 0;
> }
> EOF
>
> @@ -5322,12 +5322,16 @@ have_asan_iface_h=no
> have_asan_iface_fiber=no
>
> if test "$sanitizers" = "yes" ; then
> + if compile_prog "$CPU_CFLAGS -Werror -fsanitize=undefined" ""; then
> + have_ubsan=yes
> + fi
> +
> + # Use the skeleton for all other sanitizer checks
> + write_c_skeleton
> +
> if compile_prog "$CPU_CFLAGS -Werror -fsanitize=address" ""; then
> have_asan=yes
> fi
> - if compile_prog "$CPU_CFLAGS -Werror -fsanitize=undefined" ""; then
> - have_ubsan=yes
> - fi
>
> if check_include "sanitizer/asan_interface.h" ; then
> have_asan_iface_h=yes
> --
> 2.7.4
>
next prev parent reply other threads:[~2018-02-21 9:12 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-02-21 1:03 [Qemu-devel] [PATCH] configure: fix sanitizers' test program to mend ASan detection Emilio G. Cota
2018-02-21 9:12 ` Marc-André Lureau [this message]
2018-02-21 19:13 ` Emilio G. Cota
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=CAMxuvazD3r_-soineNOs1ZEdfUZdrYuCOq5j2xJVb0ZY5k7iaQ@mail.gmail.com \
--to=marcandre.lureau@redhat.com \
--cc=cota@braap.org \
--cc=pbonzini@redhat.com \
--cc=qemu-devel@nongnu.org \
/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 NNTP newsgroup(s).