Flashbench-results August 2014

flashbench-results@lists.linaro.org

3 participants
3 discussions

by Martin Steigerwald

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

10 years, 3 months

Samsung Memory MB-MP64D 64GB EVO MicroSDXC UHS-I Grade Class 10 MicroSDXC

by Martin Steigerwald

Hello! Three runs of my flashreport script[1] which uses flashbench as well as the open au benchmarks it suggested but additionally with report preferred erase block size as well. It seems pretty clear to me from the factory partitioning and the benchmark results that the device prefers an erase block size of large 16 MiB. But even with the reported default erase block size it seems to work quite well. Do you agree with this interpretation? I am not sure about the open au numbers as benchmark results do not vary much. It appears to me that it seems to slightly prefer lower open au numbers like 3 or 5 seem to be slightly preferable. For the intended use in Sandisk e260v1 with Rockbox open au numbers would not matter anyway as they are only setable with F2FS I think. The card indeed does work in it, reformatted to FAT32 (from ExFAT) and with partition type 0xc. I re-used the manufacturer partitioning. The included SD adapter from Samsung seems to have contact issues with SD card slot in ThinkPad T520. An Intenso I already had here one has no such issues. Except from kern.log: Aug 31 10:37:04 merkaba kernel: [215894.957529] mmc0: card 0001 removed Aug 31 10:37:06 merkaba kernel: [215896.457997] mmc0: new high speed SDXC card at address 0001 Aug 31 10:37:06 merkaba kernel: [215896.458528] mmcblk0: mmc0:0001 00000 58.5 GiB Aug 31 10:37:06 merkaba kernel: [215896.460570] mmc0: Got data interrupt 0x00200000 even though no data operation was in progress. Aug 31 10:37:06 merkaba kernel: [215896.460717] mmcblk0: error -110 sending status command, retrying Aug 31 10:37:06 merkaba kernel: [215896.460768] mmcblk0: error -110 sending status command, retrying Aug 31 10:37:06 merkaba kernel: [215896.460820] mmcblk0: error -110 sending status command, aborting Aug 31 10:37:06 merkaba kernel: [215896.460833] blk_update_request: 8 callbacks suppressed Aug 31 10:37:06 merkaba kernel: [215896.460837] end_request: I/O error, dev mmcblk0, sector 0 Aug 31 10:37:06 merkaba kernel: [215896.460842] quiet_error: 8 callbacks suppressed Aug 31 10:37:06 merkaba kernel: [215896.460846] Buffer I/O error on device mmcblk0, logical block 0 Aug 31 10:37:06 merkaba kernel: [215896.470999] mmc0: Controller never released inhibit bit(s). Aug 31 10:37:06 merkaba kernel: [215896.471207] mmcblk0: error -5 sending status command, retrying Aug 31 10:37:06 merkaba kernel: [215896.471272] mmcblk0: error -110 sending status command, retrying Aug 31 10:37:06 merkaba kernel: [215896.471324] mmcblk0: error -110 sending status command, aborting Aug 31 10:37:06 merkaba kernel: [215896.471337] end_request: I/O error, dev mmcblk0, sector 0 Aug 31 10:37:06 merkaba kernel: [215896.471343] Buffer I/O error on device mmcblk0, logical block 0 Aug 31 10:37:06 merkaba kernel: [215896.481525] mmc0: Controller never released inhibit bit(s). Aug 31 10:37:06 merkaba kernel: [215896.481678] mmcblk0: error -5 sending status command, retrying Aug 31 10:37:06 merkaba kernel: [215896.481742] mmcblk0: error -110 sending status command, retrying Aug 31 10:37:06 merkaba kernel: [215896.481796] mmcblk0: error -110 sending status command, aborting Aug 31 10:37:06 merkaba kernel: [215896.481809] end_request: I/O error, dev mmcblk0, sector 0 If I really insert it slowly and press it till the arrester it seems to work, but I rather prefer the Intenso adapter for now. Benchmark results follow: ===== Benchmarks in ThinkPad T520 internal SD card reader ===== merkaba:~> lspci -nn | grep MMC 0d:00.0 System peripheral [0880]: Ricoh Co Ltd MMC/SD Host Controller [1180:e822] (rev 08) merkaba:~> tail -fn0 /var/log/kern.log Aug 31 10:34:58 merkaba kernel: [215768.587240] mmc0: card 0001 removed Aug 31 10:35:50 merkaba kernel: [215820.887063] mmc0: new high speed SDXC card at address 0001 Aug 31 10:35:50 merkaba kernel: [215820.887507] mmcblk0: mmc0:0001 00000 58.5 GiB Aug 31 10:35:50 merkaba kernel: [215820.888892] mmcblk0: p1 Device has 62881005568 bytes: martin@merkaba:~> factor 62881005568 62881005568: 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 937 === First run of flashreport script === martin@merkaba:~/[…]/flashbench> ./flashreport /dev/mmcblk0 ===> MMC card detected with erase block size of 524288 bytes <=== ==> find /sys/block/mmcblk0/device/ -maxdepth 1 -type f | xargs head <== ==> /sys/block/mmcblk0/device/cid <== 1b534d303030303010f216606c00e300 ==> /sys/block/mmcblk0/device/csd <== 400e00325b590001d47f7f800a404000 ==> /sys/block/mmcblk0/device/scr <== 02c5800300000000 ==> /sys/block/mmcblk0/device/date <== 03/2014 ==> /sys/block/mmcblk0/device/name <== 00000 ==> /sys/block/mmcblk0/device/type <== SD ==> /sys/block/mmcblk0/device/preferred_erase_size <== 16777216 ==> /sys/block/mmcblk0/device/fwrev <== 0x0 ==> /sys/block/mmcblk0/device/hwrev <== 0x1 ==> /sys/block/mmcblk0/device/oemid <== 0x534d ==> /sys/block/mmcblk0/device/manfid <== 0x00001b ==> /sys/block/mmcblk0/device/serial <== 0xf216606c ==> /sys/block/mmcblk0/device/uevent <== DRIVER=mmcblk MMC_TYPE=SD MMC_NAME=00000 MODALIAS=mmc:block ==> /sys/block/mmcblk0/device/erase_size <== 512 ==> fdisk -lu <== [sudo] password for martin: Disk /dev/mmcblk0: 58.6 GiB, 62881005568 bytes, 122814464 sectors Units: sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disklabel type: dos Disk identifier: 0x00000000 Device Boot Start End Sectors Size Id Type /dev/mmcblk0p1 32768 122814463 122781696 58.6G 7 HPFS/NTFS/exFAT ==> sfdisk -d <== # partition table of /dev/mmcblk0 unit: sectors /dev/mmcblk0p1 : start= 32768, size=122781696, Id= 7 /dev/mmcblk0p2 : start= 0, size= 0, Id= 0 /dev/mmcblk0p3 : start= 0, size= 0, Id= 0 /dev/mmcblk0p4 : start= 0, size= 0, Id= 0 ===> Partition details <=== ==> sudo file -sk /dev/mmcblk0 <== /dev/mmcblk0: DOS/MBR boot sector; partition 1 : ID=0x7, start-CHS (0x2,10,9), end-CHS (0x3ff,254,63), startsector 32768, 122781696 sectors, extended partition table (last)\011 ==> sudo file -sk /dev/mmcblk0p1 <== /dev/mmcblk0p1: DOS/MBR boot sector ==> head -c20M | hexdump -C<== 00000000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 000001b0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 0a |................| 000001c0 09 02 07 fe ff ff 00 80 00 00 00 80 51 07 00 00 |............Q...| 000001d0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 000001f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 00000200 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 01000000 eb 76 90 45 58 46 41 54 20 20 20 00 00 00 00 00 |.v.EXFAT .....| 01000010 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 01000040 00 80 00 00 00 00 00 00 00 80 51 07 00 00 00 00 |..........Q.....| 01000050 00 40 00 00 00 40 00 00 00 80 00 00 00 51 07 00 |.@...@.......Q..| 01000060 04 00 00 00 00 00 00 00 00 01 00 00 09 08 01 80 |................| 01000070 00 00 00 00 00 00 00 00 f4 f4 f4 f4 f4 f4 f4 f4 |................| 01000080 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 |................| * 010001f0 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 55 aa |..............U.| 01000200 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 010003f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01000400 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 010005f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01000600 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 010007f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01000800 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 010009f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01000a00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 01000bf0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01000c00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 01000df0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01000e00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 01000ff0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01001000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 010011f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01001200 46 7e 0c 0a 99 33 21 40 90 c8 fa 6d 38 9c 4b a2 |F~...3!@...m8.K.| 01001210 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 01001600 d3 f8 4a 22 d3 f8 4a 22 d3 f8 4a 22 d3 f8 4a 22 |..J"..J"..J"..J"| * 01001800 eb 76 90 45 58 46 41 54 20 20 20 00 00 00 00 00 |.v.EXFAT .....| 01001810 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 01001840 00 80 00 00 00 00 00 00 00 80 51 07 00 00 00 00 |..........Q.....| 01001850 00 40 00 00 00 40 00 00 00 80 00 00 00 51 07 00 |.@...@.......Q..| 01001860 04 00 00 00 00 00 00 00 00 01 00 00 09 08 01 80 |................| 01001870 00 00 00 00 00 00 00 00 f4 f4 f4 f4 f4 f4 f4 f4 |................| 01001880 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 |................| * 010019f0 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 f4 55 aa |..............U.| 01001a00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 01001bf0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01001c00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 01001df0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01001e00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 01001ff0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01002000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 010021f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01002200 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 010023f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01002400 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 010025f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01002600 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 010027f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01002800 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 010029f0 00 00 00 00 00 00 00 00 00 00 00 00 00 00 55 aa |..............U.| 01002a00 46 7e 0c 0a 99 33 21 40 90 c8 fa 6d 38 9c 4b a2 |F~...3!@...m8.K.| 01002a10 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 01002e00 d3 f8 4a 22 d3 f8 4a 22 d3 f8 4a 22 d3 f8 4a 22 |..J"..J"..J"..J"| * 01003000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................| * 01400000 ===> Trying to find erase block size by read tests <=== ==> sudo flashbench -a --count=100 --blocksize=1024 /dev/mmcblk0 <== align 17179869184 pre 439µs on 579µs post 277µs diff 221µs align 8589934592 pre 440µs on 580µs post 277µs diff 221µs align 4294967296 pre 437µs on 578µs post 274µs diff 223µs align 2147483648 pre 440µs on 580µs post 278µs diff 221µs align 1073741824 pre 440µs on 579µs post 277µs diff 221µs align 536870912 pre 440µs on 579µs post 277µs diff 221µs align 268435456 pre 439µs on 578µs post 277µs diff 221µs align 134217728 pre 446µs on 588µs post 286µs diff 222µs align 67108864 pre 444µs on 584µs post 282µs diff 221µs align 33554432 pre 414µs on 551µs post 275µs diff 207µs align 16777216 pre 422µs on 562µs post 287µs diff 207µs align 8388608 pre 280µs on 404µs post 281µs diff 124µs align 4194304 pre 284µs on 407µs post 279µs diff 126µs align 2097152 pre 291µs on 415µs post 291µs diff 124µs align 1048576 pre 295µs on 420µs post 291µs diff 127µs align 524288 pre 293µs on 417µs post 293µs diff 124µs align 262144 pre 287µs on 411µs post 282µs diff 127µs align 131072 pre 287µs on 411µs post 283µs diff 126µs align 65536 pre 287µs on 411µs post 291µs diff 122µs align 32768 pre 274µs on 397µs post 273µs diff 123µs align 16384 pre 274µs on 398µs post 274µs diff 124µs align 8192 pre 273µs on 273µs post 273µs diff 348ns align 4096 pre 281µs on 281µs post 281µs diff 136ns align 2048 pre 267µs on 267µs post 266µs diff 557ns (See README for an explaination on how to interpret the values.) Medium is a memory card which reports 524288 bytes as erase block size and 16777216 bytes as preferred erase size. ===> Suggestions for further benchmarking <=== Further tests can show the number of open erase blocks the medium can support. BEWARE: These tests write onto the device! Make a backup first! Further tests could look as follows: flashbench --open-au --open-au-nr=3 --erasesize=524288 --blocksize=1024 /dev/mmcblk0 --offset=$[24*1024*1024] flashbench --open-au --open-au-nr=5 --erasesize=524288 --blocksize=1024 /dev/mmcblk0 --offset=$[24*1024*1024] flashbench --open-au --open-au-nr=7 --erasesize=524288 --blocksize=1024 /dev/mmcblk0 --offset=$[24*1024*1024] flashbench --open-au --open-au-nr=13 --erasesize=524288 --blocksize=1024 /dev/mmcblk0 --offset=$[24*1024*1024] (Consult the README for further suggestions.) === Second run of flashreport script === martin@merkaba:~/…/flashbench> ./flashreport /dev/mmcblk0 ===> MMC card detected with erase block size of 524288 bytes <=== […] ===> Trying to find erase block size by read tests <=== ==> sudo flashbench -a --count=100 --blocksize=1024 /dev/mmcblk0 <== align 17179869184 pre 463µs on 602µs post 301µs diff 221µs align 8589934592 pre 462µs on 601µs post 299µs diff 221µs align 4294967296 pre 448µs on 588µs post 285µs diff 221µs align 2147483648 pre 467µs on 608µs post 304µs diff 223µs align 1073741824 pre 456µs on 595µs post 293µs diff 220µs align 536870912 pre 456µs on 596µs post 292µs diff 221µs align 268435456 pre 451µs on 593µs post 286µs diff 224µs align 134217728 pre 455µs on 594µs post 288µs diff 222µs align 67108864 pre 445µs on 585µs post 284µs diff 221µs align 33554432 pre 435µs on 573µs post 293µs diff 208µs align 16777216 pre 422µs on 560µs post 284µs diff 207µs align 8388608 pre 285µs on 409µs post 285µs diff 124µs align 4194304 pre 291µs on 416µs post 287µs diff 127µs align 2097152 pre 294µs on 417µs post 293µs diff 124µs align 1048576 pre 292µs on 416µs post 288µs diff 126µs align 524288 pre 287µs on 409µs post 285µs diff 124µs align 262144 pre 283µs on 407µs post 278µs diff 126µs align 131072 pre 285µs on 409µs post 280µs diff 126µs align 65536 pre 283µs on 407µs post 287µs diff 122µs align 32768 pre 289µs on 411µs post 288µs diff 123µs align 16384 pre 278µs on 402µs post 278µs diff 124µs align 8192 pre 282µs on 281µs post 281µs diff -523ns align 4096 pre 283µs on 283µs post 283µs diff 131ns align 2048 pre 286µs on 285µs post 285µs diff -817ns (See README for an explaination on how to interpret the values.) Medium is a memory card which reports 524288 bytes as erase block size and 16777216 bytes as preferred erase size. […] === Third run of flashreport script === martin@merkaba:~/…/flashbench> ./flashreport /dev/mmcblk0 […] ===> Trying to find erase block size by read tests <=== ==> sudo flashbench -a --count=100 --blocksize=1024 /dev/mmcblk0 <== align 17179869184 pre 457µs on 597µs post 295µs diff 221µs align 8589934592 pre 453µs on 593µs post 289µs diff 222µs align 4294967296 pre 463µs on 602µs post 299µs diff 221µs align 2147483648 pre 447µs on 586µs post 284µs diff 221µs align 1073741824 pre 450µs on 590µs post 286µs diff 222µs align 536870912 pre 451µs on 590µs post 286µs diff 221µs align 268435456 pre 446µs on 585µs post 283µs diff 221µs align 134217728 pre 449µs on 589µs post 285µs diff 222µs align 67108864 pre 453µs on 594µs post 291µs diff 222µs align 33554432 pre 420µs on 557µs post 280µs diff 206µs align 16777216 pre 423µs on 560µs post 284µs diff 207µs align 8388608 pre 284µs on 408µs post 283µs diff 124µs align 4194304 pre 291µs on 415µs post 286µs diff 127µs align 2097152 pre 286µs on 409µs post 285µs diff 124µs align 1048576 pre 290µs on 414µs post 286µs diff 126µs align 524288 pre 293µs on 419µs post 293µs diff 126µs align 262144 pre 287µs on 412µs post 283µs diff 127µs align 131072 pre 285µs on 408µs post 279µs diff 126µs align 65536 pre 301µs on 426µs post 305µs diff 123µs align 32768 pre 283µs on 408µs post 283µs diff 124µs align 16384 pre 279µs on 403µs post 278µs diff 124µs align 8192 pre 279µs on 279µs post 279µs diff 300ns align 4096 pre 280µs on 279µs post 279µs diff 133ns align 2048 pre 279µs on 279µs post 279µs diff 52ns (See README for an explaination on how to interpret the values.) Medium is a memory card which reports 524288 bytes as erase block size and 16777216 bytes as preferred erase size. […] Some open au write benchmarks. It seems to me: Either the Samsung rating of 48 MB/s is highly overrated or the cardreader is not UHS-1 capable. martin@merkaba:~/[…]/flashbench> sudo flashbench --open-au --open-au-nr=3 --erasesize=524288 --blocksize=1024 /dev/mmcblk0 --offset=$[24*1024*1024] 512KiB 12.5M/s 256KiB 14M/s 128KiB 13.8M/s 64KiB 14.2M/s 32KiB 14.6M/s 16KiB 10.5M/s 8KiB 6.45M/s 4KiB 3.54M/s 2KiB 1.87M/s 1KiB 922K/s martin@merkaba:~/[…]/flashbench> sudo flashbench --open-au --open-au-nr=5 --erasesize=524288 --blocksize=1024 /dev/mmcblk0 --offset=$[24*1024*1024] 512KiB 14M/s 256KiB 14.3M/s 128KiB 12.6M/s 64KiB 13.3M/s 32KiB 12.7M/s 16KiB 11.6M/s 8KiB 7.11M/s 4KiB 3.8M/s 2KiB 1.93M/s 1KiB 973K/s martin@merkaba:~/[…]/flashbench> sudo flashbench --open-au --open-au-nr=7 --erasesize=524288 --blocksize=1024 /dev/mmcblk0 --offset=$[24*1024*1024] 512KiB 13.6M/s 256KiB 14M/s 128KiB 14.2M/s 64KiB 13.3M/s 32KiB 12.8M/s 16KiB 11.5M/s 8KiB 6.68M/s 4KiB 3.73M/s 2KiB 1.93M/s 1KiB 962K/s martin@merkaba:~/[…]/flashbench> sudo flashbench --open-au --open-au-nr=13 --erasesize=524288 --blocksize=1024 /dev/mmcblk0 --offset=$[24*1024*1024] 512KiB 13.5M/s 256KiB 14M/s 128KiB 13.9M/s 64KiB 12.9M/s 32KiB 12.4M/s 16KiB 10.7M/s 8KiB 6.79M/s 4KiB 3.76M/s 2KiB 1.86M/s 1KiB 957K/s martin@merkaba:~/[…]/flashbench> sudo flashbench --open-au --open-au-nr=3 --erasesize=16777216 --blocksize=1024 /dev/mmcblk0 --offset=$[24*1024*1024] 16MiB 14.1M/s 8MiB 13.6M/s 4MiB 13.5M/s 2MiB 13.6M/s 1MiB 13.4M/s 512KiB 13.3M/s 256KiB 13.1M/s 128KiB 12.9M/s 64KiB 12.9M/s 32KiB 13.1M/s 16KiB 10.4M/s 8KiB 6.37M/s 4KiB 3.6M/s 2KiB 1.83M/s 1KiB 948K/s martin@merkaba:~/[…]/flashbench> sudo flashbench --open-au --open-au-nr=5 --erasesize=16777216 --blocksize=1024 /dev/mmcblk0 --offset=$[24*1024*1024] 16MiB 13.8M/s 8MiB 13.6M/s 4MiB 13.5M/s 2MiB 13.7M/s 1MiB 13.5M/s 512KiB 13.3M/s 256KiB 13.4M/s 128KiB 13M/s 64KiB 12.7M/s 32KiB 12.2M/s 16KiB 9.88M/s 8KiB 6.14M/s 4KiB 3.54M/s 2KiB 1.76M/s 1KiB 897K/s martin@merkaba:~/[…]/flashbench> sudo flashbench --open-au --open-au-nr=7 --erasesize=16777216 --blocksize=1024 /dev/mmcblk0 --offset=$[24*1024*1024] 16MiB 13.7M/s 8MiB 13.6M/s 4MiB 13.6M/s 2MiB 13.4M/s 1MiB 13.4M/s 512KiB 13.1M/s 256KiB 13.3M/s 128KiB 13M/s 64KiB 12.8M/s 32KiB 12.2M/s 16KiB 9.91M/s 8KiB 6.18M/s 4KiB 3.54M/s 2KiB 1.76M/s 1KiB 914K/s martin@merkaba:~/[…]/flashbench> sudo flashbench --open-au --open-au-nr=13 --erasesize=16777216 --blocksize=1024 /dev/mmcblk0 --offset=$[24*1024*1024] 16MiB 13.9M/s 8MiB 13.6M/s 4MiB 13.8M/s 2MiB 13.4M/s 1MiB 13.5M/s 512KiB 13.2M/s 256KiB 12.8M/s 128KiB 12M/s 64KiB 11M/s 32KiB 9.37M/s 16KiB 6.45M/s 8KiB 3.07M/s 4KiB 1.6M/s 2KiB 797K/s 1KiB 408K/s ===== Benchmarks in digisol 2.0 Multi Card Reader 5.1 ===== Seems to be a Transcend reader: merkaba:~> lsusb | grep Transcend Bus 002 Device 007: ID 1307:0330 Transcend Information, Inc. 63-in-1 Multi-Card Reader/Writer merkaba:~> tail -fn0 /var/log/kern.log Aug 31 10:51:49 merkaba kernel: [216779.647323] usb 2-1.1: new high-speed USB device number 7 using ehci-pci Aug 31 10:51:49 merkaba kernel: [216779.930706] usb 2-1.1: New USB device found, idVendor=1307, idProduct=0330 Aug 31 10:51:49 merkaba kernel: [216779.930721] usb 2-1.1: New USB device strings: Mfr=1, Product=2, SerialNumber=3 Aug 31 10:51:49 merkaba kernel: [216779.930728] usb 2-1.1: Product: Mass Storage Device Aug 31 10:51:49 merkaba kernel: [216779.930734] usb 2-1.1: Manufacturer: Generic Aug 31 10:51:49 merkaba kernel: [216779.930739] usb 2-1.1: SerialNumber: 00000000000006 Aug 31 10:51:49 merkaba kernel: [216779.931712] usb-storage 2-1.1:1.0: USB Mass Storage device detected Aug 31 10:51:49 merkaba kernel: [216779.931996] scsi host13: usb-storage 2-1.1:1.0 Aug 31 10:51:50 merkaba kernel: [216780.935185] scsi 13:0:0:0: Direct-Access Generic USB CF Reader 0.00 PQ: 0 ANSI: 2 Aug 31 10:51:50 merkaba kernel: [216780.935856] scsi 13:0:0:1: Direct-Access Generic USB SD Reader 0.00 PQ: 0 ANSI: 2 Aug 31 10:51:50 merkaba kernel: [216780.936529] scsi 13:0:0:2: Direct-Access Generic Mini SD Reader 0.00 PQ: 0 ANSI: 2 Aug 31 10:51:50 merkaba kernel: [216780.937054] scsi 13:0:0:3: Direct-Access Generic USB MS Reader 0.00 PQ: 0 ANSI: 2 Aug 31 10:51:50 merkaba kernel: [216780.937556] scsi 13:0:0:4: Direct-Access Generic USB SM Reader 0.00 PQ: 0 ANSI: 2 Aug 31 10:51:50 merkaba kernel: [216780.938602] sd 13:0:0:0: Attached scsi generic sg3 type 0 Aug 31 10:51:50 merkaba kernel: [216780.940382] sd 13:0:0:1: Attached scsi generic sg4 type 0 Aug 31 10:51:50 merkaba kernel: [216780.942365] sd 13:0:0:2: Attached scsi generic sg5 type 0 Aug 31 10:51:50 merkaba kernel: [216780.945763] sd 13:0:0:0: [sdc] Attached SCSI removable disk Aug 31 10:51:50 merkaba kernel: [216780.947310] sd 13:0:0:3: Attached scsi generic sg6 type 0 Aug 31 10:51:50 merkaba kernel: [216780.948245] sd 13:0:0:4: Attached scsi generic sg7 type 0 Aug 31 10:51:50 merkaba kernel: [216780.948512] sd 13:0:0:2: [sde] Attached SCSI removable disk Aug 31 10:51:50 merkaba kernel: [216780.949139] sd 13:0:0:1: [sdd] Attached SCSI removable disk Aug 31 10:51:50 merkaba kernel: [216780.952637] sd 13:0:0:3: [sdf] Attached SCSI removable disk Aug 31 10:51:50 merkaba kernel: [216780.954517] sd 13:0:0:4: [sdg] Attached SCSI removable disk merkaba:~> tail -fn0 /var/log/kern.log Aug 31 10:54:39 merkaba kernel: [216949.712566] sd 14:0:0:1: [sdd] 2048 512-byte logical blocks: (1.04 MB/1.00 MiB) Aug 31 10:54:39 merkaba kernel: [216949.717781] sdd: sdd1 Aug 31 10:54:39 merkaba kernel: [216949.717841] sdd: p1 start 32768 is beyond EOD, enabling native capacity Aug 31 10:54:39 merkaba kernel: [216949.721094] sdd: sdd1 Aug 31 10:54:39 merkaba kernel: [216949.721129] sdd: p1 start 32768 is beyond EOD, truncated Lol, can´t read this SDXC card it seems. :) [1] git@gitorious.org:flashbench-martin21/flashbench.git Thanks, -- Martin 'Helios' Steigerwald - http://www.Lichtvoll.de GPG: 03B0 0D6C 0040 0710 4AFA B82F 991B EAAC A599 84C7

10 years, 3 months

Sandisk Extreme 16GB MicroSDHC - unclear results

by Marcel Partap

Salut, for housing a debian chroot in my android phone, I want to set up a 2.5G partition on this new micro sd card. For that, I want to figure out which parameters to give to mke2fs, as in > mkfs.ext4 -E stride=2,stripe-width=2048 -L debian_chroot /dev/sdb2 Now to figure out the internal erase block size, flashbench was recommended. But I can not quite make sense of my result.. > # fdisk.distrib -l /dev/sdb > > Disk /dev/sdb: 15.9 GB, 15931539456 bytes > 255 heads, 63 sectors/track, 1936 cylinders, total 31116288 sectors > Units = sectors of 1 * 512 = 512 bytes > Sector size (logical/physical): 512 bytes / 512 bytes > I/O size (minimum/optimal): 512 bytes / 512 bytes > Disk identifier: 0x000604f9 > > Device Boot Start End Blocks Id System > /dev/sdb1 2048 25890815 12944384 b W95 FAT32 > /dev/sdb2 25890816 31115263 2612224 83 Linux > # flashbench -va /dev/sdb --blocksize=1024 |(echo "align,pre [µs],on [µs],post [µs],diff [µs]" && sed -r 's/\w+ //g;s/\t/,/g;s/[0-9]+ns*/0/g;s/µs//g')|tee flashbench-16gb-sandisk-extreme-microsdhc-blocksize-1024.csv > # flashbench -va /dev/sdb --blocksize=$[3*1024] |(echo "align,pre [µs],on [µs],post [µs],diff [µs]" && sed -r 's/\w+ //g;s/\t/,/g;s/[0-9]+ns*/0/g;s/µs//g')|tee flashbench-16gb-sandisk-extreme-microsdhc-blocksize-$[3*1024].csv resulting CSV files attached.. I created a spreadsheet chart showing a drop.. does this mean the erasure block size is 128KB? That would give a stripe-width of 128KB/(2*4KB)=16? And I have not at all grokked the benefit/applicability of open-au/open-au-nr for practical purposes.. Any clarification greatly appreciated. #Regards!Marcel

10 years, 4 months

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

Flashbench-results August 2014