GCC segfault while compiling SPEC 2017 fprate tests

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

GCC segfault while compiling SPEC 2017 fprate tests

Steve Ellcey-9
I was doing some benchmarking with SPEC 2017 fprate on aarch64
(Thunderx2) and I am getting some segfaults from GCC while compiling.

I am working with delta to try and cut down one of the test cases
but I was wondering if anyone else has seen this problem.  The
three tests that segfault while compiling are 510.parest_r, 511.povray_r,
and 521.wrf_r.

If I compile 510.parest_r with -Ofast -std=gnu++17 -fpermissive
I get this segfault:

during GIMPLE pass: vect
source/numerics/histogram.cc: In member function 'void dealii::Histogram::evaluate(const std::vector<dealii::Vector<Number> >&, const std::vector<double>&, unsigned int, dealii::Histogram::IntervalSpacing) [with number = float]':
source/numerics/histogram.cc:54:6: internal compiler error: Segmentation fault
   54 | void Histogram::evaluate (const std::vector<Vector<number> > &values,
      |      ^~~~~~~~~
0xdfd48f crash_signal
        /home/sellcey/gcc-tot/src/gcc/gcc/toplev.c:325
0x108a6e4 contains_struct_check(tree_node*, tree_node_structure_enum, char const*, int, char const*)
        /home/sellcey/gcc-tot/src/gcc/gcc/tree.h:3231
0x108a6e4 slpeel_duplicate_current_defs_from_edges
        /home/sellcey/gcc-tot/src/gcc/gcc/tree-vect-loop-manip.c:984
0x108c87b slpeel_tree_duplicate_loop_to_edge_cfg(loop*, loop*, edge_def*)
        /home/sellcey/gcc-tot/src/gcc/gcc/tree-vect-loop-manip.c:1074
0x1090ba3 vect_do_peeling(_loop_vec_info*, tree_node*, tree_node*, tree_node**, tree_node**, tree_node**, int, bool, bool)
        /home/sellcey/gcc-tot/src/gcc/gcc/tree-vect-loop-manip.c:2580
0x108071b vect_transform_loop(_loop_vec_info*)
        /home/sellcey/gcc-tot/src/gcc/gcc/tree-vect-loop.c:8243
0x10a311f try_vectorize_loop_1
        /home/sellcey/gcc-tot/src/gcc/gcc/tree-vectorizer.c:965
0x10a3adb vectorize_loops()
        /home/sellcey/gcc-tot/src/gcc/gcc/tree-vectorizer.c:1097
Please submit a full bug report,
with preprocessed source if appropriate.
Please include the complete backtrace with any bug report.
See <https://gcc.gnu.org/bugs/> for instructions.
Reply | Threaded
Open this post in threaded view
|

Re: GCC segfault while compiling SPEC 2017 fprate tests

Arseny Solokha
> I was doing some benchmarking with SPEC 2017 fprate on aarch64
> (Thunderx2) and I am getting some segfaults from GCC while compiling.
>
> I am working with delta to try and cut down one of the test cases
> but I was wondering if anyone else has seen this problem.  The
> three tests that segfault while compiling are 510.parest_r, 511.povray_r,
> and 521.wrf_r.

This is probably PR87889, already fixed on trunk.
Reply | Threaded
Open this post in threaded view
|

Re: GCC segfault while compiling SPEC 2017 fprate tests

Richard Biener-2
In reply to this post by Steve Ellcey-9
On November 6, 2018 5:45:52 PM GMT+01:00, Steve Ellcey <[hidden email]> wrote:
>I was doing some benchmarking with SPEC 2017 fprate on aarch64
>(Thunderx2) and I am getting some segfaults from GCC while compiling.

This should be already fixed.

>I am working with delta to try and cut down one of the test cases
>but I was wondering if anyone else has seen this problem.  The
>three tests that segfault while compiling are 510.parest_r,
>511.povray_r,
>and 521.wrf_r.
>
>If I compile 510.parest_r with -Ofast -std=gnu++17 -fpermissive
>I get this segfault:
>
>during GIMPLE pass: vect
>source/numerics/histogram.cc: In member function 'void
>dealii::Histogram::evaluate(const std::vector<dealii::Vector<Number>
>>&, const std::vector<double>&, unsigned int,
>dealii::Histogram::IntervalSpacing) [with number = float]':
>source/numerics/histogram.cc:54:6: internal compiler error:
>Segmentation fault
>54 | void Histogram::evaluate (const std::vector<Vector<number> >
>&values,
>      |      ^~~~~~~~~
>0xdfd48f crash_signal
>        /home/sellcey/gcc-tot/src/gcc/gcc/toplev.c:325
>0x108a6e4 contains_struct_check(tree_node*, tree_node_structure_enum,
>char const*, int, char const*)
>        /home/sellcey/gcc-tot/src/gcc/gcc/tree.h:3231
>0x108a6e4 slpeel_duplicate_current_defs_from_edges
>        /home/sellcey/gcc-tot/src/gcc/gcc/tree-vect-loop-manip.c:984
>0x108c87b slpeel_tree_duplicate_loop_to_edge_cfg(loop*, loop*,
>edge_def*)
>        /home/sellcey/gcc-tot/src/gcc/gcc/tree-vect-loop-manip.c:1074
>0x1090ba3 vect_do_peeling(_loop_vec_info*, tree_node*, tree_node*,
>tree_node**, tree_node**, tree_node**, int, bool, bool)
>        /home/sellcey/gcc-tot/src/gcc/gcc/tree-vect-loop-manip.c:2580
>0x108071b vect_transform_loop(_loop_vec_info*)
>        /home/sellcey/gcc-tot/src/gcc/gcc/tree-vect-loop.c:8243
>0x10a311f try_vectorize_loop_1
>        /home/sellcey/gcc-tot/src/gcc/gcc/tree-vectorizer.c:965
>0x10a3adb vectorize_loops()
>        /home/sellcey/gcc-tot/src/gcc/gcc/tree-vectorizer.c:1097
>Please submit a full bug report,
>with preprocessed source if appropriate.
>Please include the complete backtrace with any bug report.
>See <https://gcc.gnu.org/bugs/> for instructions.

Reply | Threaded
Open this post in threaded view
|

Re: GCC segfault while compiling SPEC 2017 fprate tests

Steve Ellcey-9
In reply to this post by Arseny Solokha
On Wed, 2018-11-07 at 00:16 +0700, Arseny Solokha wrote:

> This is probably PR87889, already fixed on trunk.

Yup, that was the problem.  I have updated my sources and things are
building now.  Thanks for the info.

Steve Ellcey