Message ID | 1420202677-3557-1-git-send-email-will.newton@linaro.org |
---|---|
State | New |
Headers | show |
On Fri, 2 Jan 2015, Will Newton wrote: > The default required version of binutils has been increased to 2.22 > making this check redundant. OK.
On 1/2/2015 10:59 AM, Joseph Myers wrote: > On Fri, 2 Jan 2015, Will Newton wrote: > >> The default required version of binutils has been increased to 2.22 >> making this check redundant. > OK. Yes, though I'll jump in now to remind folks that I have an outstanding patch to drop the binutils minimum back to 2.21 unless there's a specific need for 2.22 somewhere. My patch doesn't affect Will's patch, but I thought I'd mention it anyway :-)
On 2 January 2015 at 16:12, Chris Metcalf <cmetcalf@ezchip.com> wrote: Hi Chris, >>> The default required version of binutils has been increased to 2.22 >>> making this check redundant. >> >> OK. > > > Yes, though I'll jump in now to remind folks that I have an outstanding > patch to drop the binutils minimum back to 2.21 unless there's a specific > need for 2.22 somewhere. My patch doesn't affect Will's patch, but I > thought I'd mention it anyway :-) Yes, it was your patch that made me check to see if there were any other checks for binutils versions in the tree. powerpc has a test for some binutils functionality but I have no idea how relevant that is now, and there was this test for ARM. As far as I am concerned moving to 2.21 is good for ARM as it lets us drop this check, but I have no opinion either way on 2.22.
On Fri, 2 Jan 2015, Chris Metcalf wrote: > On 1/2/2015 10:59 AM, Joseph Myers wrote: > > On Fri, 2 Jan 2015, Will Newton wrote: > > > > > The default required version of binutils has been increased to 2.22 > > > making this check redundant. > > OK. > > Yes, though I'll jump in now to remind folks that I have an outstanding patch > to drop the binutils minimum back to 2.21 unless there's a specific need for > 2.22 somewhere. My patch doesn't affect Will's patch, but I thought I'd > mention it anyway :-) Well, the point of 2.22 was to ensure AVX2 support; otherwise (if the copyright assignment comes through in time) the libmvec patches may need an x86_64-specific binutils version check.
On Fri, Jan 2, 2015 at 8:17 AM, Joseph Myers <joseph@codesourcery.com> wrote: > On Fri, 2 Jan 2015, Chris Metcalf wrote: > >> On 1/2/2015 10:59 AM, Joseph Myers wrote: >> > On Fri, 2 Jan 2015, Will Newton wrote: >> > >> > > The default required version of binutils has been increased to 2.22 >> > > making this check redundant. >> > OK. >> >> Yes, though I'll jump in now to remind folks that I have an outstanding patch >> to drop the binutils minimum back to 2.21 unless there's a specific need for >> 2.22 somewhere. My patch doesn't affect Will's patch, but I thought I'd >> mention it anyway :-) > > Well, the point of 2.22 was to ensure AVX2 support; otherwise (if the > copyright assignment comes through in time) the libmvec patches may need > an x86_64-specific binutils version check. I was told that Intel recently signed a blanket assignment with FSF. Can someone check it? Thanks.
On Fri, 2 Jan 2015, H.J. Lu wrote: > > Well, the point of 2.22 was to ensure AVX2 support; otherwise (if the > > copyright assignment comes through in time) the libmvec patches may need > > an x86_64-specific binutils version check. > > I was told that Intel recently signed a blanket assignment with FSF. > Can someone > check it? There is nothing in copyright.list postdating 2014-12-12.
On 1/2/2015 11:17 AM, Joseph Myers wrote: > On Fri, 2 Jan 2015, Chris Metcalf wrote: > >> On 1/2/2015 10:59 AM, Joseph Myers wrote: >>> On Fri, 2 Jan 2015, Will Newton wrote: >>> >>>> The default required version of binutils has been increased to 2.22 >>>> making this check redundant. >>> OK. >> Yes, though I'll jump in now to remind folks that I have an outstanding patch >> to drop the binutils minimum back to 2.21 unless there's a specific need for >> 2.22 somewhere. My patch doesn't affect Will's patch, but I thought I'd >> mention it anyway :-) > Well, the point of 2.22 was to ensure AVX2 support; otherwise (if the > copyright assignment comes through in time) the libmvec patches may need > an x86_64-specific binutils version check. Actually, it also seems that 2.22 is required for x32 support, so it's probably worth just admitting that if 2.22 is the right thing for x86, the rest of us can just suck it up and use --with-binutils as needed :-)
diff --git a/sysdeps/arm/armv7/configure b/sysdeps/arm/armv7/configure deleted file mode 100644 index fc8f434..0000000 --- a/sysdeps/arm/armv7/configure +++ /dev/null @@ -1,72 +0,0 @@ -# This file is generated from configure.ac by Autoconf. DO NOT EDIT! - # Local configure fragment for sysdeps/arm/armv7. - -# We need binutils 2.21 to ensure that NEON alignments are assembled correctly. -libc_cv_arm_as_version_ok=yes -for ac_prog in $AS -do - # Extract the first word of "$ac_prog", so it can be a program name with args. -set dummy $ac_prog; ac_word=$2 -{ $as_echo "$as_me:${as_lineno-$LINENO}: checking for $ac_word" >&5 -$as_echo_n "checking for $ac_word... " >&6; } -if ${ac_cv_prog_AS+:} false; then : - $as_echo_n "(cached) " >&6 -else - if test -n "$AS"; then - ac_cv_prog_AS="$AS" # Let the user override the test. -else -as_save_IFS=$IFS; IFS=$PATH_SEPARATOR -for as_dir in $PATH -do - IFS=$as_save_IFS - test -z "$as_dir" && as_dir=. - for ac_exec_ext in '' $ac_executable_extensions; do - if as_fn_executable_p "$as_dir/$ac_word$ac_exec_ext"; then - ac_cv_prog_AS="$ac_prog" - $as_echo "$as_me:${as_lineno-$LINENO}: found $as_dir/$ac_word$ac_exec_ext" >&5 - break 2 - fi -done - done -IFS=$as_save_IFS - -fi -fi -AS=$ac_cv_prog_AS -if test -n "$AS"; then - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $AS" >&5 -$as_echo "$AS" >&6; } -else - { $as_echo "$as_me:${as_lineno-$LINENO}: result: no" >&5 -$as_echo "no" >&6; } -fi - - - test -n "$AS" && break -done - -if test -z "$AS"; then - ac_verc_fail=yes -else - # Found it, now check the version. - { $as_echo "$as_me:${as_lineno-$LINENO}: checking version of $AS" >&5 -$as_echo_n "checking version of $AS... " >&6; } - ac_prog_version=`$AS --version 2>&1 | sed -n 's/^.*GNU assembler.* \([0-9]*\.[0-9.]*\).*$/\1/p'` - case $ac_prog_version in - '') ac_prog_version="v. ?.??, bad"; ac_verc_fail=yes;; - 2.1[0-9][0-9]*|2.2[1-9]*|2.[3-9][0-9]*|[3-9].*|[1-9][0-9]*) - ac_prog_version="$ac_prog_version, ok"; ac_verc_fail=no;; - *) ac_prog_version="$ac_prog_version, bad"; ac_verc_fail=yes;; - - esac - { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_prog_version" >&5 -$as_echo "$ac_prog_version" >&6; } -fi -if test $ac_verc_fail = yes; then - libc_cv_arm_as_version_ok=no -fi - - -if test $libc_cv_arm_as_version_ok != yes; then - as_fn_error $? "as version too old, at least 2.21 is required" "$LINENO" 5 -fi diff --git a/sysdeps/arm/armv7/configure.ac b/sysdeps/arm/armv7/configure.ac deleted file mode 100644 index 01e93ec..0000000 --- a/sysdeps/arm/armv7/configure.ac +++ /dev/null @@ -1,12 +0,0 @@ -GLIBC_PROVIDES dnl See aclocal.m4 in the top level source directory. -# Local configure fragment for sysdeps/arm/armv7. - -# We need binutils 2.21 to ensure that NEON alignments are assembled correctly. -libc_cv_arm_as_version_ok=yes -AC_CHECK_PROG_VER(AS, $AS, --version, - [GNU assembler.* \([0-9]*\.[0-9.]*\)], - [2.1[0-9][0-9]*|2.2[1-9]*|2.[3-9][0-9]*|[3-9].*|[1-9][0-9]*], libc_cv_arm_as_version_ok=no) - -if test $libc_cv_arm_as_version_ok != yes; then - AC_MSG_ERROR([as version too old, at least 2.21 is required]) -fi