[Bug target/104890] New: [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu

[Bug target/104890] New: [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu

[doko at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

            Bug ID: 104890
           Summary: [12 Regression] fails to build the 32bit libgcc on
                    x86_64-linux-gnu
           Product: gcc
           Version: 12.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: doko at gcc dot gnu.org
  Target Milestone: ---

trunk 20220312 fails to build the 32bit libgcc on x86_64-linux-gnu:

/home/packages/gcc/12/gcc-12-12-20220312/build/./gcc/xgcc
-B/home/packages/gcc/12/gcc-12-12-20220312/build/./gcc/ -B/usr/x86_6
4-linux-gnu/bin/ -B/usr/x86_64-linux-gnu/lib/ -isystem
/usr/x86_64-linux-gnu/include -isystem /usr/x86_64-linux-gnu/sys-includ
e -isystem /home/packages/gcc/12/gcc-12-12-20220312/build/sys-include  
-fchecking=1 -g -O2 -m32 -O2  -g -O2 -DIN_GCC    -W -W
all -Wno-narrowing -Wwrite-strings -Wcast-qual -Wstrict-prototypes
-Wmissing-prototypes -Wold-style-definition  -isystem ./inc
lude  -fpic -mlong-double-80 -DUSE_ELF_SYMVER -fcf-protection -mshstk -g
-DIN_LIBGCC2 -fbuilding-libgcc -fno-stack-protector  
-fpic -mlong-double-80 -DUSE_ELF_SYMVER -fcf-protection -mshstk -I. -I.
-I../../.././gcc -I../../../../src/libgcc -I../../../.
./src/libgcc/. -I../../../../src/libgcc/../gcc
-I../../../../src/libgcc/../include -I../../../../src/libgcc/config/libbid -DEN
ABLE_DECIMAL_BID_FORMAT -DHAVE_CC_TLS  -DUSE_TLS  -o unwind-dw2.o -MT
unwind-dw2.o -MD -MP -MF unwind-dw2.dep -fexceptions -c 
../../../../src/libgcc/unwind-dw2.c -fvisibility=hidden -DHIDE_EXPORTS
In file included from
/home/packages/gcc/12/gcc-12-12-20220312/build/gcc/include/x86gprintrin.h:45,
                 from
../../../../src/libgcc/config/i386/shadow-stack-unwind.h:25,
                 from ./md-unwind-support.h:27,
                 from ../../../../src/libgcc/unwind-dw2.c:412:
/home/packages/gcc/12/gcc-12-12-20220312/build/gcc/include/cetintrin.h: In
function '_Unwind_RaiseException':
/home/packages/gcc/12/gcc-12-12-20220312/build/gcc/include/cetintrin.h:47:1:
error: inlining failed in call to 'always_inline' '_get_ssp': target specific
option mismatch
   47 | _get_ssp (void)
      | ^~~~~~~~
../../../../src/libgcc/config/i386/shadow-stack-unwind.h:32:26: note: called
from here
   32 |       _Unwind_Word ssp = _get_ssp ();           \
      |                          ^~~~~~~~~~~
../../../../src/libgcc/unwind-dw2.c:1654:7: note: in expansion of macro
'_Unwind_Frames_Extra'
 1654 |       _Unwind_Frames_Extra (FRAMES);                                   
\
      |       ^~~~~~~~~~~~~~~~~~~~
../../../../src/libgcc/unwind.inc:140:3: note: in expansion of macro
'uw_install_context'
  140 |   uw_install_context (&this_context, &cur_context, frames);
      |   ^~~~~~~~~~~~~~~~~~
/home/packages/gcc/12/gcc-12-12-20220312/build/gcc/include/cetintrin.h:55:1:
error: inlining failed in call to 'always_inline' '_inc_ssp': target specific
option mismatch
   55 | _inc_ssp (unsigned int __B)
      | ^~~~~~~~
../../../../src/libgcc/config/i386/shadow-stack-unwind.h:38:15: note: called
from here
   38 |               _inc_ssp (255);                   \
      |               ^~~~~~~~~~~~~~
../../../../src/libgcc/unwind-dw2.c:1654:7: note: in expansion of macro
'_Unwind_Frames_Extra'
 1654 |       _Unwind_Frames_Extra (FRAMES);                                   
\
      |       ^~~~~~~~~~~~~~~~~~~~
../../../../src/libgcc/unwind.inc:140:3: note: in expansion of macro
'uw_install_context'
  140 |   uw_install_context (&this_context, &cur_context, frames);
      |   ^~~~~~~~~~~~~~~~~~
/home/packages/gcc/12/gcc-12-12-20220312/build/gcc/include/cetintrin.h:55:1:
error: inlining failed in call to 'always_inline' '_inc_ssp': target specific
option mismatch
   55 | _inc_ssp (unsigned int __B)
      | ^~~~~~~~
../../../../src/libgcc/config/i386/shadow-stack-unwind.h:41:11: note: called
from here
   41 |           _inc_ssp (tmp);                       \
      |           ^~~~~~~~~~~~~~
../../../../src/libgcc/unwind-dw2.c:1654:7: note: in expansion of macro
'_Unwind_Frames_Extra'
 1654 |       _Unwind_Frames_Extra (FRAMES);                                   
\
      |       ^~~~~~~~~~~~~~~~~~~~

../../../../src/libgcc/unwind.inc:140:3: note: in expansion of macro
'uw_install_context'
  140 |   uw_install_context (&this_context, &cur_context, frames);
      |   ^~~~~~~~~~~~~~~~~~
make[7]: *** [../../../../src/libgcc/shared-object.mk:14: unwind-dw2.o] Error 1
make[7]: Leaving directory
'/home/packages/gcc/12/gcc-12-12-20220312/build/x86_64-linux-gnu/32/libgcc'

GCC is configured with:

Configured with: -v
         --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2
         --prefix=/usr
         --with-gcc-major-version-only
         --program-suffix=-12
         --program-prefix=x86_64-linux-gnu-
         --enable-shared
         --enable-linker-build-id
         --libexecdir=/usr/lib
         --without-included-gettext
         --enable-threads=posix
         --libdir=/usr/lib
         --enable-nls
         --enable-clocale=gnu
         --enable-libstdcxx-debug
         --enable-libstdcxx-time=yes
         --with-default-libstdcxx-abi=new
         --enable-gnu-unique-object
         --disable-vtable-verify
         --enable-plugin
         --enable-default-pie
         --with-system-zlib
         --enable-libphobos-checking=release
         --with-target-system-zlib=auto
         --enable-objc-gc=auto
         --enable-multiarch
         --disable-werror
         --enable-cet
         --with-arch-32=i686
         --with-abi=m64
         --with-multilib-list=m32,m64,mx32
         --enable-multilib
         --with-tune=generic
        
--enable-offload-targets=nvptx-none=/home/packages/gcc/12/gcc-12-12-20220312/debian/tmp-nvptx/usr,amdgcn-amdhsa=/home/packages/gcc/12/gcc-12-12-20220312/debian/tmp-gcn/usr
         --enable-offload-defaulted
         --without-cuda-driver
         --enable-checking=release
         --build=x86_64-linux-gnu
         --host=x86_64-linux-gnu
         --target=x86_64-linux-gnu

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu

[doko at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

--- Comment #1 from Matthias Klose <doko at gcc dot gnu.org> ---
that is only seen when configuring with --enable-default-pie

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-default-pie)

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2022-03-12
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |WAITING

--- Comment #2 from H.J. Lu <hjl.tools at gmail dot com> ---
I can't reproduce it with

--disable-bootstrap --enable-cet --with-demangler-in-ld 
--prefix=/usr/gcc-12.0.1-x86-64-pie --with-local-prefix=/usr/local
--enable-gnu-indirect-function --enable-clocale=gnu --with-system-zlib
--with-target-system-zlib --disable-libcc1 --disable-libcilkrts
--disable-libsanitizer --enable-default-pie --enable-languages=c,c++

Please provide the prepossessed unwind-dw2.c and show the command-line
passed to cc1 with "gcc -v".

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-default-pie)

[doko at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

--- Comment #3 from Matthias Klose <doko at gcc dot gnu.org> ---
Created attachment 52616
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=52616&action=edit
preprocessed source

/home/packages/gcc/12/gcc-12-12-20220313/build/./gcc/cc1 -E -quiet -v -I . -I .
-I ../../.././gcc -I ../../../../src/libgcc -
I ../../../../src/libgcc/. -I ../../../../src/libgcc/../gcc -I
../../../../src/libgcc/../include -I ../../../../src/libgcc/con
fig/libbid -imultilib 32 -imultiarch i386-linux-gnu -iprefix
/home/packages/gcc/12/gcc-12-12-20220313/build/gcc/../lib/gcc/x86
_64-linux-gnu/12/ -isystem
/home/packages/gcc/12/gcc-12-12-20220313/build/./gcc/include -isystem
/home/packages/gcc/12/gcc-12-
12-20220313/build/./gcc/include-fixed -MD unwind-dw2.d -MF unwind-dw2.dep -MP
-MT unwind-dw2.o -D IN_GCC -D USE_ELF_SYMVER -D 
IN_LIBGCC2 -D USE_ELF_SYMVER -D ENABLE_DECIMAL_BID_FORMAT -D HAVE_CC_TLS -D
USE_TLS -D HIDE_EXPORTS -isystem /usr/x86_64-linux
-gnu/include -isystem /usr/x86_64-linux-gnu/sys-include -isystem
/home/packages/gcc/12/gcc-12-12-20220313/build/sys-include -i
system ./include ../../../../src/libgcc/unwind-dw2.c -m32 -mlong-double-80
-mshstk -mtune=generic -march=i686 -Wextra -Wall -W
no-narrowing -Wwrite-strings -Wcast-qual -Wstrict-prototypes
-Wmissing-prototypes -Wold-style-definition -fchecking=1 -fcf-pro
tection=full -fbuilding-libgcc -fno-stack-protector -fpic -fcf-protection=full
-fexceptions -fvisibility=hidden -g -g -g -fwor
king-directory -O2 -O2 -O2 -fpch-preprocess -fasynchronous-unwind-tables -o
unwind-dw2.i

In file included from
/home/packages/gcc/12/gcc-12-12-20220313/build/gcc/include/x86gprintrin.h:45,
                 from
../../../../src/libgcc/config/i386/shadow-stack-unwind.h:25,
                 from ./md-unwind-support.h:27,
                 from ../../../../src/libgcc/unwind-dw2.c:412:
/home/packages/gcc/12/gcc-12-12-20220313/build/gcc/include/cetintrin.h: In
function ‘_Unwind_RaiseException’:
/home/packages/gcc/12/gcc-12-12-20220313/build/gcc/include/cetintrin.h:47:1:
error: inlining failed in call to ‘always_inline’ ‘_get_ssp’: target specific
option mismatch
   47 | _get_ssp (void)
      | ^~~~~~~~
In file included from ../../../../src/libgcc/unwind-dw2.c:1728:
../../../../src/libgcc/unwind.inc:140:217: note: called from here
  140 |   uw_install_context (&this_context, &cur_context, frames);
      |                                                                        
                                                                               
                                                                ^          
/home/packages/gcc/12/gcc-12-12-20220313/build/gcc/include/cetintrin.h:55:1:
error: inlining failed in call to ‘always_inline’ ‘_inc_ssp’: target specific
option mismatch
   55 | _inc_ssp (unsigned int __B)
      | ^~~~~~~~
../../../../src/libgcc/unwind.inc:140:295: note: called from here
  140 |   uw_install_context (&this_context, &cur_context, frames);
      |                                                                        
                                                                               
                                                                               
                                                              ^             
/home/packages/gcc/12/gcc-12-12-20220313/build/gcc/include/cetintrin.h:55:1:
error: inlining failed in call to ‘always_inline’ ‘_inc_ssp’: target specific
option mismatch
   55 | _inc_ssp (unsigned int __B)
      | ^~~~~~~~
../../../../src/libgcc/unwind.inc:140:325: note: called from here
  140 |   uw_install_context (&this_context, &cur_context, frames);
      |                                                                        
                                                                               
                                                                               
                                                                               
            ^

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-default-pie)

[doko at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

Matthias Klose <doko at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |NEW

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-default-pie)

[doko at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

--- Comment #4 from Matthias Klose <doko at gcc dot gnu.org> ---
about the configure options, apparently pie is not the culprit, configuring
with --enable-cet is.  note that I didn't use --disable-bootstrap.

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-cet)

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at redhat dot com
             Blocks|                            |104781

--- Comment #5 from H.J. Lu <hjl.tools at gmail dot com> ---
[hjl@gnu-tgl-2 gcc]$ cat x.c 
#include <x86gprintrin.h>

__attribute__((target ("general-regs-only")))
int
foo ()
{
  return _get_ssp ();
}
[hjl@gnu-tgl-2 gcc]$ ./xgcc -B./ -S -O2 x.c -mshstk -march=i686 -m32
In file included from ./include/x86gprintrin.h:45,
                 from x.c:1:
./include/cetintrin.h: In function ‘foo’:
./include/cetintrin.h:47:1: error: inlining failed in call to ‘always_inline’
‘_get_ssp’: target specific option mismatch
   47 | _get_ssp (void)
      | ^~~~~~~~
x.c:7:10: note: called from here
    7 |   return _get_ssp ();
      |          ^~~~~~~~~~~
[hjl@gnu-tgl-2 gcc]$


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104781
[Bug 104781] [12 regression] SEGV in _Unwind_GetGR during i386 Ada bootstrap

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-cet --with-arch-32=i686)

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

--- Comment #6 from H.J. Lu <hjl.tools at gmail dot com> ---
Created attachment 52619
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=52619&action=edit
A patch

Please try this.

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-cet --with-arch-32=i686)

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |WAITING
   Target Milestone|---                         |12.0

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-cet --with-arch-32=i686)

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at gcc dot gnu.org

--- Comment #7 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Comment on attachment 52619
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=52619
A patch

This looks wrong, it will fail with the same configuration if people build with
-msse2.
The question is why __attribute__((target ("general-regs-only"))) disables
SHSTK.
If that is intentional, we should change to what my first patch proposed, i.e.
"no-sse" instead of "general-regs-only" (or "no-mmx,no-sse").
If it is unintentional, we should fix that.

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-cet --with-arch-32=i686)

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

--- Comment #8 from H.J. Lu <hjl.tools at gmail dot com> ---
(In reply to Jakub Jelinek from comment #7)
> Comment on attachment 52619 [details]
> A patch
> 
> This looks wrong, it will fail with the same configuration if people build
> with -msse2.

I have no issues to build 32-bit GCC with SSE2. I got

[hjl@gnu-tgl-2 gcc]$ ./xgcc -B./ -S -O2  -mshstk -march=i686 -m32 /tmp/x.c
In file included from ./include/x86gprintrin.h:45,
                 from /tmp/x.c:1:
./include/cetintrin.h: In function ‘foo’:
./include/cetintrin.h:47:1: error: inlining failed in call to ‘always_inline’
‘_get_ssp’: target specific option mismatch
   47 | _get_ssp (void)
      | ^~~~~~~~
/tmp/x.c:7:10: note: called from here
    7 |   return _get_ssp ();
      |          ^~~~~~~~~~~
[hjl@gnu-tgl-2 gcc]$ ./xgcc -B./ -S -O2  -mshstk -march=i686 -m32 /tmp/x.c
-msse2
[hjl@gnu-tgl-2 gcc]$

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-cet --with-arch-32=i686)

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  Attachment #52619|0                           |1
        is obsolete|                            |

--- Comment #9 from H.J. Lu <hjl.tools at gmail dot com> ---
Created attachment 52620
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=52620&action=edit
The v2 patch

Please try the v2 patch.

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-cet --with-arch-32=i686)

[doko at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

Matthias Klose <doko at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |NEW

--- Comment #10 from Matthias Klose <doko at gcc dot gnu.org> ---
The v2 patch fixes the issue for me.

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-cet --with-arch-32=i686)

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P1

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-cet --with-arch-32=i686)

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  Attachment #52620|0                           |1
        is obsolete|                            |

--- Comment #11 from H.J. Lu <hjl.tools at gmail dot com> ---
Created attachment 52625
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=52625&action=edit
The v3 patch

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-cet --with-arch-32=i686)

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

--- Comment #12 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Jakub Jelinek <jakub@gcc.gnu.org>:

https://gcc.gnu.org/g:da24fce323eaf088239e0f900dd23f6b05d552c1

commit r12-7654-gda24fce323eaf088239e0f900dd23f6b05d552c1
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Tue Mar 15 10:24:22 2022 +0100

    i386: Use no-mmx,no-sse for LIBGCC2_UNWIND_ATTRIBUTE [PR104890]

    Regardless of the outcome of the general-regs-only stuff in x86gprintrin.h,
    apparently general-regs-only is much bigger hammer than no-sse, and e.g.
    using 387 instructions in the unwinder isn't a big deal, it never needs
    to realign the stack because of it.

    So, the following patch uses no-sse (and adds no-mmx to it, even when not
    strictly needed).

    2022-03-15  Jakub Jelinek  <jakub@redhat.com>

            PR target/104890
            * config/i386/i386.h (LIBGCC2_UNWIND_ATTRIBUTE): Use no-mmx,no-sse
            instead of general-regs-only.

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-cet --with-arch-32=i686)

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

--- Comment #13 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by H.J. Lu <hjl@gcc.gnu.org>:

https://gcc.gnu.org/g:3117ffce4c1598a35e724138735b099262353358

commit r12-7673-g3117ffce4c1598a35e724138735b099262353358
Author: H.J. Lu <hjl.tools@gmail.com>
Date:   Sun Mar 13 08:57:51 2022 -0700

    x86: Also check _SOFT_FLOAT in <x86gprintrin.h>

    Push target("general-regs-only") in <x86gprintrin.h> if x87 is enabled.

    gcc/

            PR target/104890
            * config/i386/x86gprintrin.h: Also check _SOFT_FLOAT before
            pushing target("general-regs-only").

    gcc/testsuite/

            PR target/104890
            * gcc.target/i386/pr104890.c: New test.

[Bug target/104890] [12 Regression] fails to build the 32bit libgcc on x86_64-linux-gnu (--enable-cet --with-arch-32=i686)

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

--- Comment #14 from H.J. Lu <hjl.tools at gmail dot com> ---
Fixed for GCC 12 so far.

[Bug target/104890] [11 Regression] <x86gprintrin.h> doesn't work with __attribute__((target ("general-regs-only"))) and -march=i686

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |crazylht at gmail dot com

--- Comment #15 from H.J. Lu <hjl.tools at gmail dot com> ---
[hjl@gnu-skx-1 gcc]$ cat /tmp/x.c
#include <x86gprintrin.h>

__attribute__((target("no-mmx,no-sse")))
void
_Unwind_Resume_or_Rethrow(void) 
{
  _inc_ssp(1);
}
[hjl@gnu-skx-1 gcc]$ ./xgcc -B./ -O2 -m32  /tmp/x.c -S  -march=tigerlake
In file included from ./include/x86gprintrin.h:45,
                 from /tmp/x.c:1:
./include/cetintrin.h: In function ?_Unwind_Resume_or_Rethrow?:
./include/cetintrin.h:55:1: error: inlining failed in call to ?always_inline?
?_inc_ssp?: target specific option mismatch
   55 | _inc_ssp (unsigned int __B)
      | ^~~~~~~~
/tmp/x.c:7:3: note: called from here
    7 |   _inc_ssp(1);
      |   ^~~~~~~~~~~
[hjl@gnu-skx-1 gcc]$

[Bug target/104890] [11 Regression] <x86gprintrin.h> doesn't work with __attribute__((target ("general-regs-only"))) and -march=i686

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

--- Comment #16 from H.J. Lu <hjl.tools at gmail dot com> ---
[hjl@gnu-skx-1 gcc]$ ./xgcc -B./ -O2 -m32  /tmp/x.c -S  -march=sapphirerapids 
-mshstk
In file included from ./include/x86gprintrin.h:45,
                 from /tmp/x.c:1:
./include/cetintrin.h: In function ?_Unwind_Resume_or_Rethrow?:
./include/cetintrin.h:55:1: error: inlining failed in call to ?always_inline?
?_inc_ssp?: target specific option mismatch
   55 | _inc_ssp (unsigned int __B)
      | ^~~~~~~~
/tmp/x.c:7:3: note: called from here
    7 |   _inc_ssp(1);
      |   ^~~~~~~~~~~
[hjl@gnu-skx-1 gcc]$ ./xgcc -B./ -O2 -m32  /tmp/x.c -S  -march=tigerlake
-mshstkIn file included from ./include/x86gprintrin.h:45,
                 from /tmp/x.c:1:
./include/cetintrin.h: In function ?_Unwind_Resume_or_Rethrow?:
./include/cetintrin.h:55:1: error: inlining failed in call to ?always_inline?
?_inc_ssp?: target specific option mismatch
   55 | _inc_ssp (unsigned int __B)
      | ^~~~~~~~
/tmp/x.c:7:3: note: called from here
    7 |   _inc_ssp(1);
      |   ^~~~~~~~~~~
[hjl@gnu-skx-1 gcc]$ ./xgcc -B./ -O2 -m32  /tmp/x.c -S  -march=alderlake
-mshstkIn file included from ./include/x86gprintrin.h:45,
                 from /tmp/x.c:1:
./include/cetintrin.h: In function ?_Unwind_Resume_or_Rethrow?:
./include/cetintrin.h:55:1: error: inlining failed in call to ?always_inline?
?_inc_ssp?: target specific option mismatch
   55 | _inc_ssp (unsigned int __B)
      | ^~~~~~~~
/tmp/x.c:7:3: note: called from here
    7 |   _inc_ssp(1);
      |   ^~~~~~~~~~~
[hjl@gnu-skx-1 gcc]$ ./xgcc -B./ -O2 -m32  /tmp/x.c -S  -march=cascadelake
-mshstk
[hjl@gnu-skx-1 gcc]$

[Bug target/104890] [11 Regression] <x86gprintrin.h> doesn't work with __attribute__((target ("general-regs-only"))) and -march=i686

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

--- Comment #17 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The releases/gcc-11 branch has been updated by H.J. Lu <hjl@gcc.gnu.org>:

https://gcc.gnu.org/g:f407033b7ca27a75cccef8310360e275cdb94d26

commit r11-9685-gf407033b7ca27a75cccef8310360e275cdb94d26
Author: H.J. Lu <hjl.tools@gmail.com>
Date:   Sun Mar 13 08:57:51 2022 -0700

    x86: Also check _SOFT_FLOAT in <x86gprintrin.h>

    Push target("general-regs-only") in <x86gprintrin.h> if x87 is enabled.

    gcc/

            PR target/104890
            * config/i386/x86gprintrin.h: Also check _SOFT_FLOAT before
            pushing target("general-regs-only").

    gcc/testsuite/

            PR target/104890
            * gcc.target/i386/pr104890.c: New test.

    (cherry picked from commit 3117ffce4c1598a35e724138735b099262353358)

[Bug target/104890] [11 Regression] <x86gprintrin.h> doesn't work with __attribute__((target ("general-regs-only"))) and -march=i686

[hjl.tools at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=104890

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
   Target Milestone|12.0                        |11.3
             Status|NEW                         |RESOLVED

--- Comment #18 from H.J. Lu <hjl.tools at gmail dot com> ---
Fixed for GCC 12 and GCC 11.3.