问题:
RuntimeError: DataLoader worker (pid 1100) is killed by signal: Bus error.
原因:
# df -h
Filesystem Size Used Avail Use% Mounted on
overlay 3.5T 221G 3.1T 7% /
tmpfs 64M 0 64M 0% /dev
tmpfs 126G 0 126G 0% /sys/fs/cgroup
/dev/md124 500G 128G 373G 26% /headless/ld
/dev/md126 3.5T 221G 3.1T 7% /etc/hosts
shm 64M 19M 46M 30% /dev/shm ?# 这个太小了
tmpfs 126G 12K 126G 1% /proc/driver/nvidia
/dev/sda1 234G 11G 211G 5% /usr/bin/nvidia-smi
udev 126G 0 126G 0% /dev/nvidia0
tmpfs 126G 0 126G 0% /proc/asound
tmpfs 126G 0 126G 0% /proc/acpi
tmpfs 126G 0 126G 0% /proc/scsi
tmpfs 126G 0 126G 0% /sys/firmware
非docker解决: ?
sudo umount /dev/shm/
sudo mount -t tmpfs -o rw,nosuid,nodev,noexec,relatime,size=50G shm /dev/shm
docker解决:
docker run??--shm-size?50G ...
|