I am not subscribed to linaro-dev, so please Cc me in case you drop flashbench-results
mailinglist from receivers.
Hi Arnd, hi everyone!
For now quick and short.
A simple script to automate common steps on reporting flash medium test
results. It goes as far as read tests for erase block size and then suggests
some open au tests together with a warning that these write to the device.
I checked other flashbench-results posts as to what info would be
interesting.
Find it at:
git://gitorious.org/flashbench/flashbench.git
Example output is at:
[Flashbench] Samsung Plus MB-SPAGFFP Class 10 SDHC 16GB
Sat Dec 1 16:02:18 UTC 2012
http://lists.linaro.org/pipermail/flashbench-results/2012-December/000346.h…
Sorry that I occupied flashbench upstream name for project. I can look
into renaming it to "flashbench-martin" or something like that.
The script is in branch "flashreport".
In branch "defaultblocksize" you find a currently untested patch for
changing default block size to 1024.
I can try to send patches as mails via git, but I´d have to look it up first,
since I do this quite rarely.
Up to then please use the repo urls :)
Nice weekend,
--
Martin 'Helios' Steigerwald - http://www.Lichtvoll.de
GPG: 03B0 0D6C 0040 0710 4AFA B82F 991B EAAC A599 84C7
Hello,
I was testing my Kingston DataTraveler USB drive (0930:6544), for fun,
but I'm unable to interpret the following results:
$ sudo ./flashbench -a -c 64 -b 1024 /dev/sdb
align 2147483648 pre 851µs on 845µs post 852µs diff -5969ns
align 1073741824 pre 877µs on 872µs post 884µs diff -8436ns
align 536870912 pre 847µs on 833µs post 851µs diff -15795n
align 268435456 pre 851µs on 847µs post 842µs diff 787ns
align 134217728 pre 837µs on 831µs post 844µs diff -9395ns
align 67108864 pre 859µs on 823µs post 846µs diff -29289n
align 33554432 pre 841µs on 835µs post 856µs diff -13367n
align 16777216 pre 840µs on 831µs post 845µs diff -11363n
align 8388608 pre 846µs on 850µs post 850µs diff 1.9µs
align 4194304 pre 853µs on 839µs post 852µs diff -13892n
align 2097152 pre 843µs on 823µs post 849µs diff -23535n
align 1048576 pre 851µs on 852µs post 858µs diff -2358ns
align 524288 pre 855µs on 846µs post 857µs diff -9568ns
align 262144 pre 856µs on 842µs post 853µs diff -12572n
align 131072 pre 857µs on 853µs post 860µs diff -5574ns
align 65536 pre 855µs on 834µs post 842µs diff -14547n
align 32768 pre 857µs on 850µs post 851µs diff -3882ns
align 16384 pre 848µs on 846µs post 853µs diff -4037ns
align 8192 pre 866µs on 866µs post 871µs diff -2467ns
align 4096 pre 867µs on 859µs post 865µs diff -7232ns
align 2048 pre 866µs on 866µs post 868µs diff -901ns
The drive is quite old and it has been containing an ext2 filesystem
since the beginning. I only use it to mirror a directory, so it's only
plugged from time to time. But it has been almost full for a while.
Maybe the microcontroller is doing something weird, I don't really know.
Best regards.
--
Isaac Jurado
"The noblest pleasure is the joy of understanding."
Leonardo da Vinci
Looks like 4MB erase block, maybe a 128kB block and 32kb page. 7au, several
runs done as there was some variation in numbers, these are typical runs.
CompactFlash ATA device
Model Number: SanDisk SDCFH-004G
Serial Number: AAZ091113183311
Firmware Revision: HDX 7.07
pictsidhe@Miner-D:~/flashbench$ sudo ./flashbench -a /dev/sdb
--blocksize=1024
align 1073741824 pre 763µs on 980µs post 766µs diff 215µs
align 536870912 pre 884µs on 1.18ms post 888µs diff 290µs
align 268435456 pre 843µs on 1.09ms post 840µs diff 248µs
align 134217728 pre 837µs on 1.03ms post 782µs diff 223µs
align 67108864 pre 870µs on 1.12ms post 835µs diff 269µs
align 33554432 pre 810µs on 1ms post 752µs diff 222µs
align 16777216 pre 864µs on 1.15ms post 862µs diff 287µs
align 8388608 pre 802µs on 1.05ms post 807µs diff 247µs
align 4194304 pre 773µs on 1.03ms post 808µs diff 240µs
align 2097152 pre 786µs on 886µs post 785µs diff 100µs
align 1048576 pre 782µs on 909µs post 778µs diff 129µs
align 524288 pre 781µs on 885µs post 771µs diff 109µs
align 262144 pre 779µs on 886µs post 778µs diff 107µs
align 131072 pre 775µs on 882µs post 749µs diff 121µs
align 65536 pre 779µs on 882µs post 779µs diff 103µs
align 32768 pre 773µs on 895µs post 777µs diff 120µs
align 16384 pre 779µs on 848µs post 778µs diff 69.5µs
align 8192 pre 779µs on 843µs post 779µs diff 63.5µs
align 4096 pre 775µs on 859µs post 780µs diff 81µs
align 2048 pre 774µs on 843µs post 776µs diff 68.1µs
pictsidhe@Miner-D:~/flashbench$ sudo ./flashbench -O --erasesize=$[4 * 1024
* 1024] --blocksize=$[256 * 1024] /dev/sdb --open-au-nr=7
4MiB 30.1M/s
2MiB 31.3M/s
1MiB 27.1M/s
512KiB 30M/s
256KiB 29.1M/s
pictsidhe@Miner-D:~/flashbench$ sudo ./flashbench -O --erasesize=$[4 * 1024
* 1024] --blocksize=$[256 * 1024] /dev/sdb --open-au-nr=8
4MiB 31M/s
2MiB 12M/s
1MiB 3.48M/s
512KiB 1.51M/s
256KiB 703K/s
dd tests, 256MB file size:
Block: 2K 4K 8K 16K 32K 64K 128K 256K 512K 1M
Write: 1.5 MB/s 2.7 MB/s 4.9 MB/s 12.2 MB/s 18.0 MB/s 22.4 MB/s 26.0 MB/s
26.3 MB/s 27.5 MB/s 26.5 MB/s
Read: 3.7 MB/s 6.7 MB/s 11.0 MB/s 16.3 MB/s 21.5 MB/s 25.3 MB/s 27.8 MB/s
28.1 MB/s 28.2
MB/s 28.3 MB/s
--
They that can give up essential liberty to obtain a little temporary safety
deserve neither liberty nor safety. ~Benjamin Franklin
