Allow abort PLT references in libc.so for SH
Commit Message
Given my patch
<https://gcc.gnu.org/ml/gcc-patches/2017-08/msg00965.html> for the
ICEs building a glibc string function test for SH, the testsuite can
build completely for SH with GCC 7 and mainline. However, there is a
test failure that does not appear for GCC 6: check-localplt fails
because of an abort PLT reference in libc.so.
Given the lack of a trap insn pattern for SH, it seems unavoidable
that the compiler might sometimes generate abort calls, and such abort
calls (generated from __builtin_trap when there is no trap insn
pattern) will be unaffected by the normal mapping to __GI_abort for
calls within glibc. Thus, this patch allows (but does not require) an
abort PLT reference in libc.so for SH.
Tested for sh4-linux-gnu with build-many-glibcs.py (GCC 7, with my
patch applied).
2017-08-15 Joseph Myers <joseph@codesourcery.com>
* sysdeps/unix/sysv/linux/sh/localplt.data: Allow abort in
libc.so.
Comments
On 08/15/2017 05:31 PM, Joseph Myers wrote:
> Given my patch
> <https://gcc.gnu.org/ml/gcc-patches/2017-08/msg00965.html> for the
> ICEs building a glibc string function test for SH, the testsuite can
> build completely for SH with GCC 7 and mainline. However, there is a
> test failure that does not appear for GCC 6: check-localplt fails
> because of an abort PLT reference in libc.so.
>
> Given the lack of a trap insn pattern for SH, it seems unavoidable
> that the compiler might sometimes generate abort calls, and such abort
> calls (generated from __builtin_trap when there is no trap insn
> pattern) will be unaffected by the normal mapping to __GI_abort for
> calls within glibc. Thus, this patch allows (but does not require) an
> abort PLT reference in libc.so for SH.
>
> Tested for sh4-linux-gnu with build-many-glibcs.py (GCC 7, with my
> patch applied).
>
> 2017-08-15 Joseph Myers <joseph@codesourcery.com>
>
> * sysdeps/unix/sysv/linux/sh/localplt.data: Allow abort in
> libc.so.
>
> diff --git a/sysdeps/unix/sysv/linux/sh/localplt.data b/sysdeps/unix/sysv/linux/sh/localplt.data
> index 2753547..babb19d 100644
> --- a/sysdeps/unix/sysv/linux/sh/localplt.data
> +++ b/sysdeps/unix/sysv/linux/sh/localplt.data
> @@ -10,6 +10,8 @@ libc.so: _Unwind_Find_FDE
> libc.so: _exit
> libc.so: __errno_location
> libm.so: matherr
> +# Generated by the compiler because there is no trap insn pattern.
> +libc.so: abort ?
> # The main malloc is interposed into the dynamic linker, for
> # allocations after the initial link (when dlopen is used).
> ld.so: malloc
>
I agree that this is a good temporary measure, but why can't gcc be
fixed to use what we use in glibc for ABORT_INSTRUCTION?
On Wed, 16 Aug 2017, Carlos O'Donell wrote:
> I agree that this is a good temporary measure, but why can't gcc be
> fixed to use what we use in glibc for ABORT_INSTRUCTION?
See the problem in <https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70216>
agreeing *what* trapping instruction to use.
@@ -10,6 +10,8 @@ libc.so: _Unwind_Find_FDE
libc.so: _exit
libc.so: __errno_location
libm.so: matherr
+# Generated by the compiler because there is no trap insn pattern.
+libc.so: abort ?
# The main malloc is interposed into the dynamic linker, for
# allocations after the initial link (when dlopen is used).
ld.so: malloc