hi guys
seems same as class6, 16GB but it is freezing. makes the card unusable?
********************************************** **** Lexar mobile microSDHC class 4, 8 GB **** ********************************************** ==> /sys/block/mmcblk0/device/cid <== 2842454c4558415210be0131b600b500
==> /sys/block/mmcblk0/device/csd <== 400e00325b5900003bad7f800a400000
==> /sys/block/mmcblk0/device/scr <== 0235800000000000
==> /sys/block/mmcblk0/device/fwrev <== 0x0
==> /sys/block/mmcblk0/device/hwrev <== 0x1
==> /sys/block/mmcblk0/device/cid <== 2842454c4558415210be0131b600b500
==> /sys/block/mmcblk0/device/manfid <== 0x000028
==> /sys/block/mmcblk0/device/oemid <== 0x4245
==> /sys/block/mmcblk0/device/serial <== 0xbe0131b6
==> /sys/block/mmcblk0/device/erase_size <== 512
==> /sys/block/mmcblk0/device/preferred_erase_size <== 4194304
==> /sys/block/mmcblk0/device/name <== LEXAR
==> /sys/block/mmcblk0/device/date <== 05/2011 clock: 33000000 Hz vdd: 20 (3.2 ~ 3.3 V) bus mode: 2 (push-pull) chip select: 0 (don't care) power mode: 2 (on) bus width: 2 (4 bits) timing spec: 2 (sd high-speed)
$ fdisk -lu /dev/sda
Disk /dev/sda: 8010 MB, 8010072064 bytes 128 heads, 32 sectors/track, 3819 cylinders, total 15644672 sectors Units = sectors of 1 * 512 = 512 bytes Disk identifier: 0x00000000
Device Boot Start End Blocks Id System /dev/sda1 1 614399 307199+ 83 Linux /dev/sda2 614400 819199 102400 83 Linux /dev/sda3 819200 1851391 516096 82 Linux swap / Solaris /dev/sda4 1851392 7364607 2756608 83 Linux
$ ./flashbench -a /dev/sda3 --count=100 --blocksize=2048 align 134217728 pre 738µs on 968µs post 863µs diff 168µs align 67108864 pre 742µs on 973µs post 868µs diff 168µs align 33554432 pre 737µs on 970µs post 864µs diff 170µs align 16777216 pre 741µs on 974µs post 859µs diff 174µs align 8388608 pre 742µs on 973µs post 863µs diff 171µs align 4194304 pre 738µs on 975µs post 868µs diff 172µs align 2097152 pre 743µs on 855µs post 852µs diff 57.2µs align 1048576 pre 741µs on 854µs post 867µs diff 49.7µs align 524288 pre 739µs on 864µs post 867µs diff 60.4µs align 262144 pre 738µs on 866µs post 867µs diff 63.4µs align 131072 pre 737µs on 863µs post 868µs diff 60.9µs align 65536 pre 733µs on 882µs post 869µs diff 81.6µs align 32768 pre 730µs on 870µs post 865µs diff 73µs align 16384 pre 737µs on 886µs post 869µs diff 83µs align 8192 pre 734µs on 872µs post 868µs diff 70.4µs align 4096 pre 733µs on 872µs post 864µs diff 73.4µs
-> 4MiB erase-block
$ ./flashbench /dev/sda3 --open-au-nr=1 -O --blocksize=2048 4MiB 11.1M/s 2MiB 11.1M/s 1MiB 11.4M/s 512KiB 10.5M/s 256KiB 11M/s 128KiB 10.9M/s 64KiB 11.5M/s 32KiB 9.48M/s 16KiB 7.71M/s 8KiB 4.66M/s 4KiB 2.54M/s 2KiB 1.47M/s
$ ./flashbench /dev/sda3 --open-au-nr=5 -O --blocksize=2048 4MiB 11.1M/s 2MiB 5.86M/s 1MiB 11.1M/s 512KiB 11.3M/s 256KiB 11.2M/s 128KiB 10.6M/s 64KiB 10.9M/s 32KiB 9.4M/s 16KiB 7.54M/s 8KiB 4.57M/s 4KiB 2.34M/s 2KiB 1.19M/s
$ ./flashbench /dev/sda3 --open-au-nr=10 -O --blocksize=2048 4MiB 9.56M/s 2MiB 3.2M/s 1MiB 1.8M/s 512KiB 986K/s 256KiB 516K/s 128KiB 252K/s freezing, also 2nd time (!!!)
$ ./flashbench /dev/sda3 --open-au-nr=6 -O --blocksize=2048 4MiB 7.93M/s 2MiB 3.27M/s 1MiB 1.8M/s 512KiB 988K/s 256KiB 517K/s 128KiB 264K/s 64KiB 137K/s 32KiB 68.7K/s freezes (!!!)
-> can handle 5 blocks -> what fishy is going on that it freezes with more? -> is this card useful at all if it behaves like this?
$ ./flashbench /dev/sda3 --open-au-nr=5 -O --blocksize=2048 4MiB 10.1M/s 2MiB 10.4M/s 1MiB 10.5M/s 512KiB 10.3M/s 256KiB 10.1M/s 128KiB 9.51M/s 64KiB 9.54M/s 32KiB 7.66M/s 16KiB 5.48M/s 8KiB 3.17M/s 4KiB 1.64M/s 2KiB 901K/s
$ ./flashbench /dev/sda3 --open-au-nr=5 -O --blocksize=2048 --random 4MiB 10.7M/s 2MiB 10.5M/s 1MiB 10.5M/s 512KiB 10.2M/s 256KiB 9.75M/s 128KiB 9.67M/s 64KiB 9.54M/s 32KiB 7.59M/s 16KiB 5.5M/s 8KiB 3.19M/s 4KiB 1.25M/s 2KiB 613K/s
-> random is also good, same cutoff
peter
On Wednesday 13 July 2011, Peter Warasin wrote:
$ ./flashbench /dev/sda3 --open-au-nr=10 -O --blocksize=2048 4MiB 9.56M/s 2MiB 3.2M/s 1MiB 1.8M/s 512KiB 986K/s 256KiB 516K/s 128KiB 252K/s freezing, also 2nd time (!!!)
$ ./flashbench /dev/sda3 --open-au-nr=6 -O --blocksize=2048 4MiB 7.93M/s 2MiB 3.27M/s 1MiB 1.8M/s 512KiB 988K/s 256KiB 517K/s 128KiB 264K/s 64KiB 137K/s 32KiB 68.7K/s freezes (!!!)
-> can handle 5 blocks -> what fishy is going on that it freezes with more? -> is this card useful at all if it behaves like this?
Depends. How do you know that it freezes? Sometimes these tests can take a *very* long time. Do you see any messages in 'dmesg'?
If the card is a fake one, it's possible that it's already hitting some broken erase blocks.
Arnd
On 14/07/11 13:27, Arnd Bergmann wrote:
Depends. How do you know that it freezes? Sometimes these tests can take a *very* long time. Do you see any messages in 'dmesg'?
well, i aborted after a 30min of no change. on dmesg there's nothing
If the card is a fake one, it's possible that it's already hitting some broken erase blocks.
ok, should'nt i see then something in dmesg like with mtd devices? or is the block-layer hiding it?
peter
On Monday 18 July 2011, Peter Warasin wrote:
On 14/07/11 13:27, Arnd Bergmann wrote:
Depends. How do you know that it freezes? Sometimes these tests can take a *very* long time. Do you see any messages in 'dmesg'?
well, i aborted after a 30min of no change. on dmesg there's nothing
If the card is a fake one, it's possible that it's already hitting some broken erase blocks.
ok, should'nt i see then something in dmesg like with mtd devices? or is the block-layer hiding it?
You should definitely see something in dmesg if there is a hardware problem.
When you see a stall like that again, another idea is to attach an 'strace' to the flashbench process and see if it's still doing system calls.
Arnd
flashbench-results@lists.linaro.org