[Bug lto/94157] New: [10 Regression] error: lto-wrapper failed with -Wa,--noexecstack -Wa,--noexecstack

[Bug lto/94157] New: [10 Regression] error: lto-wrapper failed with -Wa,--noexecstack -Wa,--noexecstack

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94157

            Bug ID: 94157
           Summary: [10 Regression] error: lto-wrapper failed with
                    -Wa,--noexecstack  -Wa,--noexecstack
           Product: gcc
           Version: 9.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: lto
          Assignee: unassigned at gcc dot gnu.org
          Reporter: marxin at gcc dot gnu.org
                CC: marxin at gcc dot gnu.org
  Target Milestone: ---

Using gcc10 package I see a strange error:

$ echo "int main() {}" > x.c
$ gcc -flto -c x.c
$ gcc -Wa,--noexecstack  -Wa,--noexecstack   x.o
/usr/lib64/gcc/x86_64-suse-linux/10/../../../../x86_64-suse-linux/bin/ld:
error: lto-wrapper failed
collect2: error: ld returned 1 exit status
$ gcc -Wa,--noexecstack x.o
[fine]

I can't reproduce that with locally built GCC and gcc9 package seems fine.

[Bug lto/94157] [10 Regression] error: lto-wrapper failed with -Wa,--noexecstack -Wa,--noexecstack

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94157

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Assignee|unassigned at gcc dot gnu.org      |marxin at gcc dot gnu.org
             Status|UNCONFIRMED                 |ASSIGNED
     Ever confirmed|0                           |1
   Last reconfirmed|                            |2020-03-12

--- Comment #1 from Martin Liška <marxin at gcc dot gnu.org> ---
Ok, I can reproduce that now..

[Bug lto/94157] [10 Regression] error: lto-wrapper failed with -Wa,--noexecstack -Wa,--noexecstack since r10-6807-gf1a681a174cdfb82

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94157

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[10 Regression] error:      |[10 Regression] error:
                   |lto-wrapper failed with     |lto-wrapper failed with
                   |-Wa,--noexecstack           |-Wa,--noexecstack
                   |-Wa,--noexecstack           |-Wa,--noexecstack since
                   |                            |r10-6807-gf1a681a174cdfb82

--- Comment #2 from Martin Liška <marxin at gcc dot gnu.org> ---
Started with r10-6807-gf1a681a174cdfb82. One can reproduce with:

$ gcc foo.c -c -flto && gcc -fPIE -Wa,--noexecstack -Wa,--noexecstack foo.o -o
a.out -Wa,--execstack  -Wa,--execstack -Wa,--execstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack
/usr/bin/ld: error: lto-wrapper failed
collect2: error: ld returned 1 exit status

==14211== Invalid write of size 1
==14211==    at 0x483B98C: strcat (in
/usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==14211==    by 0x408A08: run_gcc(unsigned int, char**) (lto-wrapper.c:1320)
==14211==    by 0x4057C5: main (lto-wrapper.c:1981)
==14211==  Address 0x1fff001000 is not stack'd, malloc'd or (recently) free'd

[Bug lto/94157] [10 Regression] error: lto-wrapper failed with -Wa,--noexecstack -Wa,--noexecstack since r10-6807-gf1a681a174cdfb82

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94157

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |10.0
            Version|9.0                         |10.0
                 CC|                            |prathamesh3492 at gcc dot gnu.org
           Priority|P3                          |P1

[Bug lto/94157] [10 Regression] error: lto-wrapper failed with -Wa,--noexecstack -Wa,--noexecstack since r10-6807-gf1a681a174cdfb82

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94157

--- Comment #3 from Martin Liška <marxin at gcc dot gnu.org> ---
I've got a patch candidate, will send it to GCC patches mailing list.

[Bug lto/94157] [10 Regression] error: lto-wrapper failed with -Wa,--noexecstack -Wa,--noexecstack since r10-6807-gf1a681a174cdfb82

[prathamesh3492 at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94157

--- Comment #4 from prathamesh3492 at gcc dot gnu.org ---
(In reply to Martin Liška from comment #3)
> I've got a patch candidate, will send it to GCC patches mailing list.

Sorry for the breakage, and thanks for taking a look!

Regards,
Prathamesh

[Bug lto/94157] [10 Regression] error: lto-wrapper failed with -Wa,--noexecstack -Wa,--noexecstack since r10-6807-gf1a681a174cdfb82

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94157

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|ASSIGNED                    |RESOLVED

--- Comment #5 from Martin Liška <marxin at gcc dot gnu.org> ---
commit r10-7157-g98ff89d1ac5872f29020fe5b5edfdf5abce59014
Author: Martin Liska <mliska@suse.cz>
Date:   Fri Mar 13 13:50:01 2020 +0100

    Do not strcat to result of getenv.

            PR lto/94157
            * lto-wrapper.c (run_gcc): Use concat for appending
            to collect_gcc_options.
            PR lto/94157
            * gcc.dg/lto/pr94157_0.c: New test.

[Bug lto/94157] [10 Regression] error: lto-wrapper failed with -Wa,--noexecstack -Wa,--noexecstack since r10-6807-gf1a681a174cdfb82

[ro at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94157

Rainer Orth <ro at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
                 CC|                            |ro at gcc dot gnu.org
         Resolution|FIXED                       |---

--- Comment #6 from Rainer Orth <ro at gcc dot gnu.org> ---
The new test FAILs on (at least) Solaris with the native assembler (and
probably
others like Darwin and HP-UX):

+FAIL: gcc.dg/lto/pr94157 c_lto_pr94157_0.o assemble,  -O0 -fipa-vrp -flto
-Wa,--noexecstack -Wa,--noexecstack -Wa,--execstack  -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack 
+UNRESOLVED: gcc.dg/lto/pr94157 c_lto_pr94157_0.o-c_lto_pr94157_0.o execute 
-O0 -fipa-vrp -flto -Wa,--noexecstack -Wa,--noexecstack -Wa,--execstack 
-Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack 
+UNRESOLVED: gcc.dg/lto/pr94157 c_lto_pr94157_0.o-c_lto_pr94157_0.o link  -O0
-fipa-vrp -flto -Wa,--noexecstack -Wa,--noexecstack -Wa,--execstack 
-Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack 

/usr/ccs/bin/as: error: unknown option '-'^M
/usr/ccs/bin/as: error: unknown option '-'^M
/usr/ccs/bin/as: error: unknown option 'e'^M
/usr/ccs/bin/as: error: unknown option 'xecstack'^M
/usr/ccs/bin/as: error: unknown option '-'^M
/usr/ccs/bin/as: error: unknown option 'e'^M
/usr/ccs/bin/as: error: unknown option 'xecstack'^M
[...]

Either use something portable here or restrict the test to assemblers supporing
--execstack/--noexecstack.

[Bug lto/94157] [10 Regression] error: lto-wrapper failed with -Wa,--noexecstack -Wa,--noexecstack since r10-6807-gf1a681a174cdfb82

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94157

--- Comment #7 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Martin Liska <marxin@gcc.gnu.org>:

https://gcc.gnu.org/g:ecf2b69a629d4f79efe3c103fe54040437ea18a6

commit r10-7205-gecf2b69a629d4f79efe3c103fe54040437ea18a6
Author: Martin Liska <mliska@suse.cz>
Date:   Tue Mar 17 09:43:46 2020 +0100

    Filter a test-case with gas.

            PR lto/94157
            * gcc.dg/lto/pr94157_0.c: Add gas effective
            target filter.

[Bug lto/94157] [10 Regression] error: lto-wrapper failed with -Wa,--noexecstack -Wa,--noexecstack since r10-6807-gf1a681a174cdfb82

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94157

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|REOPENED                    |RESOLVED
         Resolution|---                         |FIXED

--- Comment #8 from Martin Liška <marxin at gcc dot gnu.org> ---
Fixed.

[Bug lto/94157] [10 Regression] error: lto-wrapper failed with -Wa,--noexecstack -Wa,--noexecstack since r10-6807-gf1a681a174cdfb82

[jiawei at iscas dot ac.cn]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94157

jiawei <jiawei at iscas dot ac.cn> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jiawei at iscas dot ac.cn

--- Comment #9 from jiawei <jiawei at iscas dot ac.cn> ---
I had run the regression on the RISC-V target, and get a ld warning:

`requires executable stack (because the .note.GNU-stack section is executable)`

Executing on host: /root/riscv-gnu-toolchain/build-gcc-newlib-stage2/gcc/xgcc
-B/root/riscv-gnu-toolchain/build-gcc-newlib-stage2/gcc/ c_lto_pr94157_0.o 
-march=rv64imafdc -mabi=lp64d -mcmodel=medlow   -fdiagnostics-plain-output  
-O0 -fipa-vrp -flto -Wa,--noexecstack -Wa,--noexecstack -Wa,--execstack 
-Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack          -o
gcc-dg-lto-pr94157-01.exe    (timeout = 600)
spawn -ignore SIGHUP /root/riscv-gnu-toolchain/build-gcc-newlib-stage2/gcc/xgcc
-B/root/riscv-gnu-toolchain/build-gcc-newlib-stage2/gcc/ c_lto_pr94157_0.o
-march=rv64imafdc -mabi=lp64d -mcmodel=medlow -fdiagnostics-plain-output -O0
-fipa-vrp -flto -Wa,--noexecstack -Wa,--noexecstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack -o gcc-dg-lto-pr94157-01.exe^M
/opt/riscv/riscv64-unknown-elf/bin/ld: warning: /tmp/ccrEUuLH.ltrans0.ltrans.o:
requires executable stack (because the .note.GNU-stack section is executable)^M
FAIL: gcc.dg/lto/pr94157 c_lto_pr94157_0.o-c_lto_pr94157_0.o link,  -O0
-fipa-vrp -flto -Wa,--noexecstack -Wa,--noexecstack -Wa,--execstack 
-Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack -Wa,--execstack
-Wa,--execstack -Wa,--execstack -Wa,--execstack

Any suggestions?

[Bug lto/94157] [10 Regression] error: lto-wrapper failed with -Wa,--noexecstack -Wa,--noexecstack since r10-6807-gf1a681a174cdfb82

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94157

--- Comment #10 from Martin Liška <marxin at gcc dot gnu.org> ---
Seems you are using the latest binutils ld, right?

It's the newly added warning which tells that usage of executable stack is a
potential security issue:
https://sourceware.org/pipermail/binutils/2022-April/120476.html

[Bug lto/94157] [10 Regression] error: lto-wrapper failed with -Wa,--noexecstack -Wa,--noexecstack since r10-6807-gf1a681a174cdfb82

[jiawei at iscas dot ac.cn]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94157

--- Comment #11 from jiawei <jiawei at iscas dot ac.cn> ---
(In reply to Martin Liška from comment #10)
> Seems you are using the latest binutils ld, right?
> 
> It's the newly added warning which tells that usage of executable stack is a
> potential security issue:
> https://sourceware.org/pipermail/binutils/2022-April/120476.html

Yes, you are all right, so it seems not an error, thanks a lot!!