/dev/mmcblk1p1:
Timing cached reads: 1504 MB in 2.00 seconds = 751.48 MB/sec
Timing buffered disk reads: 60 MB in 3.02 seconds = 19.85 MB/sec
osmc@osmc-QuantumRay:~$ sudo hdparm -Tt /dev/mmcblk1p1
/dev/mmcblk1p1:
Timing cached reads: 1784 MB in 2.00 seconds = 892.10 MB/sec
Timing buffered disk reads: 56 MB in 3.02 seconds = 18.57 MB/sec
osmc@osmc-QuantumRay:~$ sudo hdparm -Tt /dev/mmcblk1p1
/dev/mmcblk1p1:
Timing cached reads: 1740 MB in 2.00 seconds = 869.74 MB/sec
Timing buffered disk reads: 58 MB in 3.01 seconds = 19.26 MB/sec
This is sequential read performance
fio --name TEST --eta-newline=5s --filename=DarkSaber/fio-tempfile.dat --rw=read --size=500m --io_size=10g --blocksize=1024k --ioengine=libaio --fsync=10000 --iodepth=32 --direct=1 --numjobs=1 --runtime=60 --group_reporting
Run status group 0 (all jobs):
** READ: bw=20.0MiB/s (21.0MB/s), 20.0MiB/s-20.0MiB/s (21.0MB/s-21.0MB/s), io=1276MiB (1338MB), run=60864-60864msec**
This is sequential write performance
fio --name TEST --eta-newline=5s --filename=DarkSaber/fio-tempfile.dat --rw=write --size=500m --io_size=10g --blocksize=1024k --ioengine=libaio --fsync=10000 --iodepth=32 --direct=1 --numjobs=1 --runtime=60 --group_reporting
Run status group 0 (all jobs):
** WRITE: bw=17.3MiB/s (18.2MB/s), 17.3MiB/s-17.3MiB/s (18.2MB/s-18.2MB/s), io=1056MiB (1107MB), run=60926-60926msec**
I certainly did not expect for write to be worse?
I checked the reader configuration
osmc@osmc-QuantumRay:~$ sudo cat /sys/kernel/debug/emmc/ios
clock: 100000000 Hz
actual clock: 100000000 Hz
vdd: 21 (3.3 ~ 3.4 V)
bus mode: 2 (push-pull)
chip select: 0 (don’t care)
power mode: 2 (on)
bus width: 3 (8 bits)
timing spec: 10 (mmc HS400)
signal voltage: 1 (1.80 V)
driver type: 0 (driver type B)
It seems this mode MMC HS400 have a maximum transfer rate of 400 MB/s and 200 MHz SDCLK frequency. SDCLK frequency is 100 MHz not sure what else can be affecting it.
Have you tried it in another device?
Are you using WiFi?
No, the wireless adapter is disabled. The torrent service and the SSH is running. I was testing with all torrents stopped. I will try to test on my Razer hub reader. But unless the card is a brutal fake, I doubt it will show similar speeds.
I did a few tests on my laptop. The reader is Genesys Logic model attached via Thunderbolt 4
The results are what you can expect from this card
READ
TEST: (groupid=0, jobs=1): err= 0: pid=2388655: Thu Apr 20 23:17:05 2023
read: IOPS=174, BW=174MiB/s (183MB/s)(10.0GiB/58682msec)
slat (usec): min=19, max=28267, avg=5701.69, stdev=4663.43
clat (msec): min=11, max=203, avg=172.34, stdev=24.44
lat (msec): min=11, max=206, avg=178.04, stdev=24.48
WRITE
TEST: (groupid=0, jobs=1): err= 0: pid=2409862: Thu Apr 20 23:19:09 2023
write: IOPS=134, BW=134MiB/s (141MB/s)(8053MiB/60008msec); 0 zone resets
slat (usec): min=39, max=71295, avg=7430.19, stdev=5227.67
clat (msec): min=5, max=300, avg=222.93, stdev=35.45
lat (msec): min=12, max=312, avg=230.36, stdev=35.36
I also corrected the device I am using, its Vero 4k +
Are you sure you tested this with the same program?
The result output looks very different.
I assume this is also a Linux laptop?
Yes, its the same tool - fio, using exactly the same command line, although it might be different version as my laptop is Fedora 37 and uses different repos than Vero.