Flashbench-results

flashbench-results@lists.linaro.org

8 participants
745 discussions

Re: [Flashbench] flashbench : time_read: Invalid argument

by arnd.bergmann＠linaro.org

On Saturday 05 March 2011 00:55:29 Philippe De Muyter wrote: > I have read with interest your article in lwn, and decided to try your flashbench > program to discover the characteristics of the CF card we use. > > Unfortunately, > > ./flashbench -a -b 1K /dev/sdb > > fails with endless : > > time_read: Invalid argument > > Looking with strace, I get e.g.: > > pread64(3, 0xb3728000, 1, 1023410175) = -1 EINVAL (Invalid argument) > > Is that a behavior you can explain (and fix) ? Yes, there are two known problems that I need to fix: 1. flashbench uses O_DIRECT for talking to the device, which means that all accesses must be on multiples of full sectors based on the underlying device sector size. It should detect this 2. command line arguments currently need to be natural numbers. There is no parser for interpreting arguments like 1K or 4M as kilobytes and megabytes. It should work when you do ./flashbench -a -b 1024 /dev/sdb Arnd

15 years, 2 months

Re: [Flashbench] flashbench: Usage and understanding -s and -f ?

by Arnd Bergmann

On Sunday 13 March 2011, Dirk Behme wrote: > Hi Arnd, > > reading the flashbench README [1] what I really like is the > explanation of the -a and the -O option. Explaining the options in > this README, giving the examples and how to interpret the resulting > figures really does help. > > Somehow I miss something similar for the -s and -f options. I.e. how > to select the proper values for --scatter-order and --scatter-span and > how to interpret the output of -s and -f. > > Once I understood it, I would be able to send a patch for the README ;) > > Additionally, it would be nice to give the flashbench options used for > the graphs [2] [3] in the LWN article. The LWN article explains quite > nicely how to interpret the given graphs, but it's not mentioned which > flashbench options were used to get these graphs. > > Many thanks for your help and best regards > > Dirk > > [1] > http://git.linaro.org/gitweb?p=people/arnd/flashbench.git;a=blob;f=README;h… > Sorry for the late reply. I promise I'll get to it and update the README. I should actually remove --scatter-order, it's too difficult to understand this. It specifies the log2 of the number of blocks in terms of --blocksize to be tested at the start of the medium. The output file can be interpreted by gnuplot -p -e 'plot "output.file"' or by importing it into a spreadsheet program like oocalc and using the XY chart function on two columns. For --findfat, the output shows how each of the first N erase blocks on the drive reacts to certain access patterns within the erase block. Most drives do something different for a few blocks in the beginning to optimize storing the FAT on them. Each column is one erase block here. If they are all the same, the card does not have an optimzied FAT area. Arnd

15 years, 2 months

Re: [Flashbench] Fwd: flashbench results APACER SD Embedded(SLC) 4GB:

by Arnd Bergmann

On Friday 04 March 2011 20:16:05 Xianghua Xiao wrote: > From the data below it appears APACER has a 1MB erasing block instead > of the typical 4MB? Yes, and this is not surprising for an SLC card, although it is the first one I have seen. > Also from the following email(waiting for approval on the list, will > forward to you soon), that looks like a 4MB erasing block but is > slower than APACER. Sorry, I need to find the place on the mailing list settings to allow non-members to post, and approve the mails you already sent. > Does open-au-nr with higher number mean the underlying filesystem esp > ext4 will do better, Yes > does that imply multithread parallel writing to some extent? No, it's not about threads, but about how the data is laid out. ext4 will have to write data, metadata and journal data for many accesses, but these three are normally in different locations on the driver, so you need at least three open segments for the first process that is writing data. If you have other processes that also write to the drive, or one process writing to multiple files, you will need more. > open-au-nr means how many different segments you can use > to write to in the same time(so you don't need wait for one segment)? Yes. The apacer card evidently can write to three segments, but not to six. Can you also try 4 and 5 segments? The interesting number is the maximum. For the unigen card, it looks like it does not handle random access well, independent of the number of AUs. Best try again without --random. > I probably will recommend APACER over UNIGEN, does that make sense? Depends on the other measurements. Arnd

15 years, 3 months

flashbench result: Unigen SDHC 4GB

by Xianghua Xiao

./flashbench -a /dev/sdd2 align 67108864 pre 1.6ms on 1.6ms post 1.52ms diff 34.9µs align 33554432 pre 1.59ms on 1.59ms post 1.53ms diff 31.6µs align 16777216 pre 1.58ms on 1.59ms post 1.52ms diff 36.8µs align 8388608 pre 1.71ms on 1.72ms post 1.6ms diff 70.6µs align 4194304 pre 1.72ms on 1.72ms post 1.58ms diff 71.5µs align 2097152 pre 1.54ms on 1.55ms post 1.51ms diff 21.1µs align 1048576 pre 1.55ms on 1.57ms post 1.52ms diff 30.6µs align 524288 pre 1.55ms on 1.56ms post 1.52ms diff 25µs align 262144 pre 1.52ms on 1.52ms post 1.51ms diff 2.85µs align 131072 pre 1.51ms on 1.52ms post 1.52ms diff 3.14µs align 65536 pre 1.52ms on 1.52ms post 1.52ms diff 244ns align 32768 pre 1.52ms on 1.52ms post 1.52ms diff 4.51µs ./flashbench --open-au --open-au-nr=1 /dev/sdd2 4MiB 8.34M/s 2MiB 8.57M/s 1MiB 8.71M/s 512KiB 8.74M/s 256KiB 8.57M/s 128KiB 7.96M/s 64KiB 16.9M/s 32KiB 5.89M/s 16KiB 3.56M/s ./flashbench --open-au --open-au-nr=2 /dev/sdd2 4MiB 8.69M/s 2MiB 7.47M/s 1MiB 6.12M/s 512KiB 8.79M/s 256KiB 8.52M/s 128KiB 8.22M/s 64KiB 8.57M/s 32KiB 5.99M/s 16KiB 3.66M/s ./flashbench --open-au --open-au-nr=3 /dev/sdd2 4MiB 8.51M/s 2MiB 7.48M/s 1MiB 6.04M/s 512KiB 4.37M/s 256KiB 2.43M/s 128KiB 1.42M/s 64KiB 808K/s 32KiB 425K/s 16KiB 216K/s head /sys/block/sdd/device/* ==> /sys/block/sdd/device/block <== head: error reading `/sys/block/sdd/device/block': Is a directory ==> /sys/block/sdd/device/bsg <== head: error reading `/sys/block/sdd/device/bsg': Is a directory head: cannot open `/sys/block/sdd/device/delete' for reading: Permission denied ==> /sys/block/sdd/device/device_blocked <== 0 ==> /sys/block/sdd/device/dh_state <== detached ==> /sys/block/sdd/device/driver <== head: error reading `/sys/block/sdd/device/driver': Is a directory ==> /sys/block/sdd/device/evt_media_change <== 0 ==> /sys/block/sdd/device/generic <== head: error reading `/sys/block/sdd/device/generic': Is a directory ==> /sys/block/sdd/device/iocounterbits <== 32 ==> /sys/block/sdd/device/iodone_cnt <== 0x249d5d ==> /sys/block/sdd/device/ioerr_cnt <== 0x218be3 ==> /sys/block/sdd/device/iorequest_cnt <== 0x249d5d ==> /sys/block/sdd/device/max_sectors <== 240 ==> /sys/block/sdd/device/modalias <== scsi:t-0x00 ==> /sys/block/sdd/device/model <== STORAGE DEVICE ==> /sys/block/sdd/device/power <== head: error reading `/sys/block/sdd/device/power': Is a directory ==> /sys/block/sdd/device/queue_depth <== 1 ==> /sys/block/sdd/device/queue_type <== none head: cannot open `/sys/block/sdd/device/rescan' for reading: Permission denied ==> /sys/block/sdd/device/rev <== 9744 ==> /sys/block/sdd/device/scsi_device <== head: error reading `/sys/block/sdd/device/scsi_device': Is a directory ==> /sys/block/sdd/device/scsi_disk <== head: error reading `/sys/block/sdd/device/scsi_disk': Is a directory ==> /sys/block/sdd/device/scsi_generic <== head: error reading `/sys/block/sdd/device/scsi_generic': Is a directory ==> /sys/block/sdd/device/scsi_level <== 0 ==> /sys/block/sdd/device/state <== running ==> /sys/block/sdd/device/subsystem <== head: error reading `/sys/block/sdd/device/subsystem': Is a directory ==> /sys/block/sdd/device/timeout <== 30 ==> /sys/block/sdd/device/type <== 0 ==> /sys/block/sdd/device/uevent <== DEVTYPE=scsi_device DRIVER=sd MODALIAS=scsi:t-0x00 ==> /sys/block/sdd/device/vendor <== Generic

15 years, 3 months

flashbench results APACER SD Embedded(SLC) 4GB:

by Xianghua Xiao

APACER SD Embedded(SLC) 4GB: ./flashbench -a /dev/sdd2 align 67108864 pre 1.97ms on 1.96ms post 1.61ms diff 164µs align 33554432 pre 1.95ms on 1.95ms post 1.63ms diff 163µs align 16777216 pre 1.95ms on 1.94ms post 1.63ms diff 147µs align 8388608 pre 1.96ms on 1.96ms post 1.63ms diff 164µs align 4194304 pre 1.94ms on 1.94ms post 1.63ms diff 161µs align 2097152 pre 1.95ms on 1.94ms post 1.63ms diff 149µs align 1048576 pre 1.93ms on 1.94ms post 1.63ms diff 157µs align 524288 pre 1.62ms on 1.63ms post 1.62ms diff 2.07µs align 262144 pre 1.62ms on 1.63ms post 1.63ms diff 398ns align 131072 pre 1.62ms on 1.63ms post 1.63ms diff -620ns align 65536 pre 1.63ms on 1.63ms post 1.63ms diff 530ns align 32768 pre 1.62ms on 1.63ms post 1.63ms diff 2.37µs ./flashbench --open-au --open-au-nr=1 /dev/sdd2 4MiB 11.1M/s 2MiB 8.64M/s 1MiB 10M/s 512KiB 11.4M/s 256KiB 11.4M/s 128KiB 11.1M/s 64KiB 17.8M/s 32KiB 9.01M/s 16KiB 5.99M/s ./flashbench --open-au --open-au-nr=2 /dev/sdd2 4MiB 6.78M/s 2MiB 4.58M/s 1MiB 2.62M/s 512KiB 4.78M/s 256KiB 7M/s 128KiB 9.18M/s 64KiB 9.69M/s 32KiB 7.76M/s 16KiB 4.97M/s ./flashbench --open-au --open-au-nr=3 /dev/sdd2 4MiB 5.53M/s 2MiB 4.04M/s 1MiB 2.62M/s 512KiB 5.04M/s 256KiB 7.41M/s 128KiB 7.15M/s 64KiB 7.26M/s 32KiB 5.21M/s 16KiB 3.04M/s head /sys/block/sdd/device/* ==> /sys/block/sdd/device/block <== head: error reading `/sys/block/sdd/device/block': Is a directory ==> /sys/block/sdd/device/bsg <== head: error reading `/sys/block/sdd/device/bsg': Is a directory head: cannot open `/sys/block/sdd/device/delete' for reading: Permission denied ==> /sys/block/sdd/device/device_blocked <== 0 ==> /sys/block/sdd/device/dh_state <== detached ==> /sys/block/sdd/device/driver <== head: error reading `/sys/block/sdd/device/driver': Is a directory ==> /sys/block/sdd/device/evt_media_change <== 0 ==> /sys/block/sdd/device/generic <== head: error reading `/sys/block/sdd/device/generic': Is a directory ==> /sys/block/sdd/device/iocounterbits <== 32 ==> /sys/block/sdd/device/iodone_cnt <== 0x243f01 ==> /sys/block/sdd/device/ioerr_cnt <== 0x218ba6 ==> /sys/block/sdd/device/iorequest_cnt <== 0x243f01 ==> /sys/block/sdd/device/max_sectors <== 240 ==> /sys/block/sdd/device/modalias <== scsi:t-0x00 ==> /sys/block/sdd/device/model <== STORAGE DEVICE ==> /sys/block/sdd/device/power <== head: error reading `/sys/block/sdd/device/power': Is a directory ==> /sys/block/sdd/device/queue_depth <== 1 ==> /sys/block/sdd/device/queue_type <== none head: cannot open `/sys/block/sdd/device/rescan' for reading: Permission denied ==> /sys/block/sdd/device/rev <== 9744 ==> /sys/block/sdd/device/scsi_device <== head: error reading `/sys/block/sdd/device/scsi_device': Is a directory ==> /sys/block/sdd/device/scsi_disk <== head: error reading `/sys/block/sdd/device/scsi_disk': Is a directory ==> /sys/block/sdd/device/scsi_generic <== head: error reading `/sys/block/sdd/device/scsi_generic': Is a directory ==> /sys/block/sdd/device/scsi_level <== 0 ==> /sys/block/sdd/device/state <== running ==> /sys/block/sdd/device/subsystem <== head: error reading `/sys/block/sdd/device/subsystem': Is a directory ==> /sys/block/sdd/device/timeout <== 30 ==> /sys/block/sdd/device/type <== 0 ==> /sys/block/sdd/device/uevent <== DEVTYPE=scsi_device DRIVER=sd MODALIAS=scsi:t-0x00 ==> /sys/block/sdd/device/vendor <== Generic

15 years, 3 months

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

Flashbench-results