From: "H.J. Lu" <hjl.tools@gmail.com>
To: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Cc: GNU C Library <libc-alpha@sourceware.org>
Subject: Re: [PATCH] Check alignment of PT_LOAD segment only if possible
Date: Fri, 10 Dec 2021 13:54:05 -0800 [thread overview]
Message-ID: <CAMe9rOqP6SJ7WsoYYqGXhENyCZ23G4WS9TJnQXsHyKTmqvkBFQ@mail.gmail.com> (raw)
In-Reply-To: <edf7cf1b-bcd3-aa35-241a-039aabae9e49@linaro.org>
On Fri, Dec 10, 2021 at 1:36 PM Adhemerval Zanella
<adhemerval.zanella@linaro.org> wrote:
>
>
>
> On 10/12/2021 17:59, H.J. Lu wrote:
> > Add a configure check for 2MB variable alignment and check alignment of
> > PT_LOAD segment only if 2MB variable alignment is supported.
> >
> > Tested with cross build to glibcs-alpha-linux-gnu.
> > ---
> > configure | 40 +++++++++++++++++++++++++++++++++++++++-
> > configure.ac | 11 +++++++++++
> > elf/Makefile | 15 ++++++++++-----
> > 3 files changed, 60 insertions(+), 6 deletions(-)
> >
> > diff --git a/configure b/configure
> > index e4d013907c..9da30f428f 100755
> > --- a/configure
> > +++ b/configure
> > @@ -732,6 +732,7 @@ infodir
> > docdir
> > oldincludedir
> > includedir
> > +runstatedir
> > localstatedir
> > sharedstatedir
> > sysconfdir
> > @@ -845,6 +846,7 @@ datadir='${datarootdir}'
> > sysconfdir='${prefix}/etc'
> > sharedstatedir='${prefix}/com'
> > localstatedir='${prefix}/var'
> > +runstatedir='${localstatedir}/run'
> > includedir='${prefix}/include'
> > oldincludedir='/usr/include'
> > docdir='${datarootdir}/doc/${PACKAGE_TARNAME}'
> > @@ -1097,6 +1099,15 @@ do
> > | -silent | --silent | --silen | --sile | --sil)
> > silent=yes ;;
> >
> > + -runstatedir | --runstatedir | --runstatedi | --runstated \
> > + | --runstate | --runstat | --runsta | --runst | --runs \
> > + | --run | --ru | --r)
> > + ac_prev=runstatedir ;;
> > + -runstatedir=* | --runstatedir=* | --runstatedi=* | --runstated=* \
> > + | --runstate=* | --runstat=* | --runsta=* | --runst=* | --runs=* \
> > + | --run=* | --ru=* | --r=*)
> > + runstatedir=$ac_optarg ;;
> > +
> > -sbindir | --sbindir | --sbindi | --sbind | --sbin | --sbi | --sb)
> > ac_prev=sbindir ;;
> > -sbindir=* | --sbindir=* | --sbindi=* | --sbind=* | --sbin=* \
>
> Please use a non patched autoconf (usually debian based one adds the runstatedir).
Fixed.
> > @@ -1234,7 +1245,7 @@ fi
> > for ac_var in exec_prefix prefix bindir sbindir libexecdir datarootdir \
> > datadir sysconfdir sharedstatedir localstatedir includedir \
> > oldincludedir docdir infodir htmldir dvidir pdfdir psdir \
> > - libdir localedir mandir
> > + libdir localedir mandir runstatedir
> > do
> > eval ac_val=\$$ac_var
> > # Remove trailing slashes.
> > @@ -1387,6 +1398,7 @@ Fine tuning of the installation directories:
> > --sysconfdir=DIR read-only single-machine data [PREFIX/etc]
> > --sharedstatedir=DIR modifiable architecture-independent data [PREFIX/com]
> > --localstatedir=DIR modifiable single-machine data [PREFIX/var]
> > + --runstatedir=DIR modifiable per-process data [LOCALSTATEDIR/run]
> > --libdir=DIR object code libraries [EPREFIX/lib]
> > --includedir=DIR C header files [PREFIX/include]
> > --oldincludedir=DIR C header files for non-gcc [/usr/include]
> > @@ -6817,6 +6829,32 @@ if test $libc_cv_builtin_trap = yes; then
> >
> > fi
> >
> > +# Check target support for 2MB variable alignment.
> > +{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for 2MB variable alignment" >&5
> > +$as_echo_n "checking for 2MB variable alignment... " >&6; }
> > +if ${libc_cv_have_2MB_alignment+:} false; then :
> > + $as_echo_n "(cached) " >&6
> > +else
> > +
> > +cat confdefs.h - <<_ACEOF >conftest.$ac_ext
> > +/* end confdefs.h. */
> > +
> > +int foo __attribute__((aligned(0x200000))) = 1;
> > +
> > +_ACEOF
> > +if ac_fn_c_try_compile "$LINENO"; then :
> > + libc_cv_have_2MB_alignment=yes
> > +else
> > + libc_cv_have_2MB_alignment=no
> > +fi
> > +rm -f core conftest.err conftest.$ac_objext conftest.$ac_ext
> > +
> > +fi
> > +{ $as_echo "$as_me:${as_lineno-$LINENO}: result: $libc_cv_have_2MB_alignment" >&5
> > +$as_echo "$libc_cv_have_2MB_alignment" >&6; }
> > +config_vars="$config_vars
> > +have-2MB-alignment = $libc_cv_have_2MB_alignment"
> > +
> > ac_ext=cpp
> > ac_cpp='$CXXCPP $CPPFLAGS'
> > ac_compile='$CXX -c $CXXFLAGS $CPPFLAGS conftest.$ac_ext >&5'
> > diff --git a/configure.ac b/configure.ac
> > index a91a7f399c..57d4572c26 100644
> > --- a/configure.ac
> > +++ b/configure.ac
> > @@ -1726,6 +1726,17 @@ if test $libc_cv_builtin_trap = yes; then
> > AC_DEFINE([HAVE_BUILTIN_TRAP])
> > fi
> >
> > +# Check target support for 2MB variable alignment.
> > +AC_CACHE_CHECK([for 2MB variable alignment],
> > + libc_cv_have_2MB_alignment, [
> > +AC_COMPILE_IFELSE([AC_LANG_SOURCE([
> > +int foo __attribute__((aligned(0x200000))) = 1;
> > +])],
> > + [libc_cv_have_2MB_alignment=yes],
> > + [libc_cv_have_2MB_alignment=no])
> > +])
> > +LIBC_CONFIG_VAR([have-2MB-alignment], [$libc_cv_have_2MB_alignment])
> > +
> > dnl C++ feature tests.
> > AC_LANG_PUSH([C++])
> >
>
> Would be better to check 2 times the maximum supported pagesize instead?
Why 2 times? The test only uses 2MB alignment. The implementation
self doesn't depend on the maximum supported page size. The limit
is available free address space.
> > diff --git a/elf/Makefile b/elf/Makefile
> > index fe42caeb0e..1bf77bc5ac 100644
> > --- a/elf/Makefile
> > +++ b/elf/Makefile
> > @@ -207,7 +207,7 @@ tests += restest1 preloadtest loadfail multiload origtest resolvfail \
> > tst-tls4 tst-tls5 \
> > tst-tls10 tst-tls11 tst-tls12 tst-tls13 tst-tls14 tst-tls15 \
> > tst-tls16 tst-tls17 tst-tls18 tst-tls19 tst-tls-dlinfo \
> > - tst-align tst-align2 tst-align3 \
> > + tst-align tst-align2 \
> > tst-dlmodcount tst-dlopenrpath tst-deep1 \
> > tst-dlmopen1 tst-dlmopen3 tst-dlmopen4 \
> > unload3 unload4 unload5 unload6 unload7 unload8 tst-global1 order2 \
> > @@ -241,9 +241,6 @@ tests-internal += loadtest unload unload2 circleload1 \
> > tests-container += tst-pldd tst-dlopen-tlsmodid-container \
> > tst-dlopen-self-container tst-preload-pthread-libc
> > test-srcs = tst-pathopt
> > -ifeq (yes,$(have-fpie))
> > -tests-pie += tst-align3
> > -endif
> > selinux-enabled := $(shell cat /selinux/enforce 2> /dev/null)
> > ifneq ($(selinux-enabled),1)
> > tests-execstack-yes = tst-execstack tst-execstack-needed tst-execstack-prog
> > @@ -305,7 +302,7 @@ modules-names = testobj1 testobj2 testobj3 testobj4 testobj5 testobj6 \
> > circlemod3 circlemod3a \
> > reldep8mod1 reldep8mod2 reldep8mod3 \
> > reldep9mod1 reldep9mod2 reldep9mod3 \
> > - tst-alignmod tst-alignmod2 tst-alignmod3 \
> > + tst-alignmod tst-alignmod2 \
> > $(modules-execstack-$(have-z-execstack)) \
> > tst-dlopenrpathmod tst-deep1mod1 tst-deep1mod2 tst-deep1mod3 \
> > tst-dlmopen1mod tst-auditmod1 \
> > @@ -418,6 +415,14 @@ endif
> > modules-execstack-yes = tst-execstack-mod
> > extra-test-objs += $(addsuffix .os,$(strip $(modules-names)))
> >
> > +ifeq (yes,$(have-2MB-alignment))
> > +tests += tst-align3
> > +modules-names += tst-alignmod3
> > +ifeq (yes,$(have-fpie))
> > +tests-pie += tst-align3
> > +endif
> > +endif
> > +
> > # filtmod1.so, tst-big-note-lib.so, tst-ro-dynamic-mod.so have special
> > # rules.
> > modules-names-nobuild := filtmod1 tst-big-note-lib tst-ro-dynamic-mod
> >
--
H.J.
next prev parent reply other threads:[~2021-12-10 21:54 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-12-10 20:59 H.J. Lu
2021-12-10 21:36 ` Adhemerval Zanella
2021-12-10 21:54 ` H.J. Lu [this message]
2021-12-10 22:50 ` 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=CAMe9rOqP6SJ7WsoYYqGXhENyCZ23G4WS9TJnQXsHyKTmqvkBFQ@mail.gmail.com \
--to=hjl.tools@gmail.com \
--cc=adhemerval.zanella@linaro.org \
--cc=libc-alpha@sourceware.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 read-only IMAP folder(s) and NNTP newsgroup(s).