Slow USB on my old RaspPi ?

[tomcheng76]

This is the test, the usb harddrive is Seagate expansion desktop
$this->bbcode_second_pass_code('', 'dd if=/dev/zero of=/media/external/test.img bs=2k count=1024k')
On Goflex Net , it has 27MiB/s, on Respberry Pi, it has only 12-15MiB/s
Is it Normal? The RaspPi is old 256M Model B, latest alarmpi image with pacman -Syu
Please report your speed
Thanks

[sdjf]

Not able to test right now, but I believe the class of your SD card plays a major role in speed. I am using class 4 and know that is slower than the class 10 cards, from what I have read.

[tomcheng76]

I am not sure if SD card is related, since pi is writing to the separated hard drive, anyway it is a class 6 SD card.

Here are the mount entries, notice that the seagate drive is mount as ext3 (rw,noatime,data=writeback)
$this->bbcode_second_pass_code('', '/dev/mmcblk0p5 on / type ext4 (rw,noatime,discard,data=ordered)
devtmpfs on /dev type devtmpfs (rw,relatime,size=51560k,nr_inodes=12890,mode=755)
proc on /proc type proc (rw,nosuid,nodev,noexec,relatime)
sysfs on /sys type sysfs (rw,nosuid,nodev,noexec,relatime)
securityfs on /sys/kernel/security type securityfs (rw,nosuid,nodev,noexec,relatime)
tmpfs on /dev/shm type tmpfs (rw,nosuid,nodev)
devpts on /dev/pts type devpts (rw,nosuid,noexec,relatime,gid=5,mode=620)
tmpfs on /run type tmpfs (rw,nosuid,nodev,mode=755)
tmpfs on /sys/fs/cgroup type tmpfs (rw,nosuid,nodev,noexec,mode=755)
cgroup on /sys/fs/cgroup/systemd type cgroup (rw,nosuid,nodev,noexec,relatime,xattr,release_agent=/usr/lib/systemd/systemd-cgroups-agent,name=systemd)
cgroup on /sys/fs/cgroup/cpuset type cgroup (rw,nosuid,nodev,noexec,relatime,cpuset)
cgroup on /sys/fs/cgroup/cpu,cpuacct type cgroup (rw,nosuid,nodev,noexec,relatime,cpuacct,cpu)
cgroup on /sys/fs/cgroup/devices type cgroup (rw,nosuid,nodev,noexec,relatime,devices)
cgroup on /sys/fs/cgroup/freezer type cgroup (rw,nosuid,nodev,noexec,relatime,freezer)
cgroup on /sys/fs/cgroup/blkio type cgroup (rw,nosuid,nodev,noexec,relatime,blkio)
cgroup on /sys/fs/cgroup/perf_event type cgroup (rw,nosuid,nodev,noexec,relatime,perf_event)
cgroup on /sys/fs/cgroup/net_prio type cgroup (rw,nosuid,nodev,noexec,relatime,net_prio)
systemd-1 on /proc/sys/fs/binfmt_misc type autofs (rw,relatime,fd=22,pgrp=1,timeout=300,minproto=5,maxproto=5,direct)
mqueue on /dev/mqueue type mqueue (rw,relatime)
debugfs on /sys/kernel/debug type debugfs (rw,relatime)
configfs on /sys/kernel/config type configfs (rw,relatime)
tmpfs on /var/log type tmpfs (rw,nosuid,nodev,noexec,noatime,size=4096k,mode=755)
tmpfs on /tmp type tmpfs (rw,nosuid,nodev,relatime)
/dev/sda1 on /media/external type ext3 (rw,noatime,data=writeback)
/dev/mmcblk0p1 on /boot type vfat (rw,relatime,fmask=0022,dmask=0022,codepage=437,iocharset=ascii,shortname=mixed,errors=remount-ro)
')

[moonman]

Remember that this is OTG USB port and the same host is shared between multiple devices, including the builtin network adapter.

[tomcheng76]

$this->bbcode_second_pass_quote('moonman', 'R')emember that this is OTG USB port and the same host is shared between multiple devices, including the builtin network adapter.

[moonman]

I don't think you can. In theory it samba should be ~15 MB/s since USB is maxed out at 30, and you are using NIC on the same USB controller as the hard drive.

[tomcheng76]

Thanks for the explanation , new config get 6-7MiB/s, it is just a little bit sad that i cant get 10MiB/s

$this->bbcode_second_pass_code('', ' socket options = TCP_NODELAY IPTOS_LOWDELAY SO_SNDBUF=65535 SO_RCVBUF=65535
max connections = 65535
max open files = 65535')

[moonman]

Check the cpuload and iowait in top while doing the transfer. Also try the -latest kernel with elevator=bfq in cmdline.txt. It might help

[tomcheng76]

Here is the top snapshot
$this->bbcode_second_pass_code('', 'top - 18:38:57 up 22:08, 1 user, load average: 1.10, 0.42, 0.29
Tasks: 75 total, 3 running, 72 sleeping, 0 stopped, 0 zombie
%Cpu(s): 5.8 us, 40.8 sy, 0.0 ni, 4.9 id, 0.6 wa, 0.6 hi, 47.2 si, 0.0 st
KiB Mem: 213952 total, 58408 used, 155544 free, 1568 buffers
KiB Swap: 524284 total, 27152 used, 497132 free. 16836 cached Mem

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
19295 bomber 20 0 42620 1752 864 R 66.8 0.8 5:50.54 smbd
3 root 20 0 0 0 0 R 6.1 0.0 29:36.29 ksoftirqd/0
60 root 20 0 0 0 0 D 4.2 0.0 18:43.08 usb-storage
2896 http 20 0 30240 5288 1748 D 3.8 2.5 31:14.40 php-fpm
20 root 20 0 0 0 0 S 2.9 0.0 12:13.24 kswapd0
144 root 20 0 4808 240 240 S 2.6 0.1 13:14.18 haveged
19477 root 20 0 0 0 0 S 1.9 0.0 0:02.02 kworker/u2:2
46 root 20 0 0 0 0 S 1.3 0.0 14:59.81 mmcqd/0
140 root 20 0 1672 184 184 S 1.3 0.1 5:36.52 rngd
19750 bomber 20 0 4764 1228 952 R 1.0 0.6 0:00.24 top
287 http 20 0 5728 924 688 S 0.6 0.4 1:27.97 nginx
1 root 20 0 4216 1752 1164 S 0.3 0.8 0:52.67 systemd
47 root 0 -20 0 0 0 S 0.3 0.0 0:56.06 kworker/0:1H
145 dbus 20 0 3632 708 564 S 0.3 0.3 0:29.87 dbus-daemon
19054 root 20 0 0 0 0 S 0.3 0.0 0:24.99 kworker/0:2
2 root 20 0 0 0 0 S 0.0 0.0 0:00.07 kthreadd
5 root 0 -20 0 0 0 S 0.0 0.0 0:00.00 kworker/0:0H')

Thanks for the idea!
It is now running bfq, so i think there is not much i can do, nearly no swap activity

[moonman]

Cpu is maxed out and soft interrupts are eating 47%. Basically cpu spend half of its time handling interrupts and this is due to the driver. You may try to disable fiq split transaction but it may cause other errors like jittery sound.