Created by: sgrvinod

I'm trying to use the pretrained wmt14 translation model from the demo. It works fine on the CPU, but as soon as I set --use_gpu=1, it gives the following error:

I0222 16:08:09.291610 30305 Util.cpp:155] commandline: /usr/bin/../opt/paddle/bin/paddle_trainer --job=test --config=translation/gen.conf --save_dir=data/wmt14_model --use_gpu=1 --num_passes=13 --test_pass=12 --trainer_count=1 --config_args=beam_size=3,max_length=100 
*** stack smashing detected ***: /usr/bin/../opt/paddle/bin/paddle_trainer terminated
*** Aborted at 1487808490 (unix time) try "date -d @1487808490" if you are using GNU date ***
PC: @     0x7f81132fcc37 (unknown)
*** SIGABRT (@0x7661) received by PID 30305 (TID 0x7f8115609840) from PID 30305; stack trace: ***
    @     0x7f8114f05330 (unknown)
    @     0x7f81132fcc37 (unknown)
    @     0x7f8113300028 (unknown)
    @     0x7f81133392a4 (unknown)
    @     0x7f81133d0bbc (unknown)
    @     0x7f81133d0b60 (unknown)
    @           0x8954ce hl_create_global_resources()
    @           0x8959f9 hl_specify_devices_start()
    @           0x895d2d hl_start()
    @           0x818602 paddle::initMain()
    @           0x52ac5b main
    @     0x7f81132e7f45 (unknown)
    @           0x540c05 (unknown)
    @                0x0 (unknown)
/usr/bin/paddle: line 109: 30305 Aborted                 (core dumped) ${DEBUGGER} $MYDIR/../opt/paddle/bin/paddle_trainer ${@:2}

My NVIDIA-SMI, if useful:

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 367.48                 Driver Version: 367.48                    |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  TITAN X (Pascal)    Off  | 0000:02:00.0     Off |                  N/A |
| 23%   38C    P8    15W / 250W |      1MiB / 12189MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
|   1  TITAN X (Pascal)    Off  | 0000:03:00.0     Off |                  N/A |
| 23%   37C    P8    16W / 250W |      1MiB / 12189MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
|   2  TITAN X (Pascal)    Off  | 0000:04:00.0     Off |                  N/A |
| 23%   33C    P8     8W / 250W |      1MiB / 12189MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
|   3  TITAN X (Pascal)    Off  | 0000:05:00.0      On |                  N/A |
| 28%   47C    P8    17W / 250W |    460MiB / 12188MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+

I'm using

PaddlePaddle 0.9.0, compiled with
    with_avx: ON
    with_gpu: ON
    with_double: OFF
    with_python: ON
    with_rdma: OFF
    with_glog: ON
    with_gflags: ON
    with_metric_learning: 
    with_timer: OFF
    with_predict_sdk: 

Cuda V8.0.44 Ubuntu 14.04

Any help would be appreciated, thanks!