SD Card UHS speed

[hopkinskong]

Does RockPro64 support UHS speed?
It seems that not many SBC have support the high speed SD bus due to the lower voltage requirement requires extra hardware to detect the type of card to switch to lower voltage and directly using lower voltage will breaking backward compatibility.

However, for a competent hardware like RockPro64 (2Ghz CPU, PCI-E, etc.), not having a fast SD card bus seems a bit of shame.

[Luke]

Does RockPro64 support UHS speed?
It seems that not many SBC have support the high speed SD bus due to the lower voltage requirement requires extra hardware to detect the type of card to switch to lower voltage and directly using lower voltage will breaking backward compatibility.

However, for a competent hardware like RockPro64 (2Ghz CPU, PCI-E, etc.), not having a fast SD card bus seems a bit of shame.

I honestly cannot see a reason for SDs anymore with a board like the RPro64. You can use eMMC, USB 3.0 / C drives (remember SPI boot is an option) and even other types of drives via PCIe ...
Sorry for not answering your question (I don't know the answer) but I feel a SD would only be useful for testing an image on a board like this.

[xalius]

Excluding any hardware issues, UHS-I modes should work since the regulator used to supply the port for the mmc bus and the card can switch between 1.8V and 3.3V... as far as I have seen reports from other RK3399 boards the software support is also in place in the BSP kernel at least...

[z4v4l]

Could anybody owning the board and a UHS-I capable SD card, show iozone test results for them? Please.

I want to get the board, I like it from all the rk3399 boards most, but I only can buy if it does support UHS-I.

[dukla2000]

Could anybody owning the board and a UHS-I capable SD card, show iozone test results for them? Please.

Makes sense to me to use an NVMe as wipes the floor with eMMC or SDcard. Still have my SDcard in for boot so

Code:

$ uname -a
Linux rpro64.dukla.net 4.4.138-1100-rockchip-ayufan-g95cecee47f40 #1 SMP Sat Sep 29 15:43:04 UTC 2018 aarch64 aarch64 aarch64 GNU/Linux
$ iozone -e -I -a -s 10M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2
    Iozone: Performance Test of File I/O
            Version $Revision: 3.429 $
        Compiled for 64 bit mode.
        Build: linux

    Contributors:William Norcott, Don Capps, Isom Crawford, Kirby Collins
                 Al Slater, Scott Rhine, Mike Wisner, Ken Goss
                 Steve Landherr, Brad Smith, Mark Kelly, Dr. Alain CYR,
                 Randy Dunlap, Mark Montague, Dan Million, Gavin Brebner,
                 Jean-Marc Zucconi, Jeff Blomberg, Benny Halevy, Dave Boone,
                 Erik Habbinga, Kris Strecker, Walter Wong, Joshua Root,
                 Fabrice Bacchella, Zhenghua Xue, Qin Li, Darren Sawyer,
                 Vangel Bojaxhi, Ben England, Vikentsi Lapa.

    Run began: Thu Dec  6 21:26:27 2018

    Include fsync in write timing
    O_DIRECT feature enabled
    Auto Mode
    File size set to 10240 kB
    Record Size 4 kB
    Record Size 16 kB
    Record Size 512 kB
    Record Size 1024 kB
    Record Size 16384 kB
    Command line used: iozone -e -I -a -s 10M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2
    Output is in kBytes/sec
    Time Resolution = 0.000001 seconds.
    Processor cache size set to 1024 kBytes.
    Processor cache line size set to 32 bytes.
    File stride size set to 17 * record size.
                                                              random    random     bkwd    record    stride                                    
              kB  reclen    write  rewrite    read    reread    read     write     read   rewrite      read   fwrite frewrite    fread  freread
           10240       4     2085     1404     8684     8760     8742      784                                                          
           10240      16     5121     6378    16028    15916    16022    11567                                                          
           10240     512    15535    15499    20967    21248    21280    16811                                                          
           10240    1024    18969    19248    22551    22666    22667    19433                                                          

iozone test complete.

Note I cut the filesize to 10M for the SDcard - 100M numbers for NVMe are here.

Will update this post with 4.19 numbers after a reboot and will also confirm the SDcard spec.

PS - OK, my SDcard is Samsung EVO Plus 128Gb. As expected 4.19 significantly faster than 4.4,138-1100 I used above (loads of noise on these forums elsewhere why).

Code:

$ uname -a
Linux rpro64.dukla.net 4.19.0 #4 SMP PREEMPT Sun Dec 2 23:57:11 GMT 2018 aarch64 aarch64 aarch64 GNU/Linux
$ iozone -e -I -a -s 10M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2
    Iozone: Performance Test of File I/O
            Version $Revision: 3.429 $
        Compiled for 64 bit mode.
        Build: linux

    Contributors:William Norcott, Don Capps, Isom Crawford, Kirby Collins
                 Al Slater, Scott Rhine, Mike Wisner, Ken Goss
                 Steve Landherr, Brad Smith, Mark Kelly, Dr. Alain CYR,
                 Randy Dunlap, Mark Montague, Dan Million, Gavin Brebner,
                 Jean-Marc Zucconi, Jeff Blomberg, Benny Halevy, Dave Boone,
                 Erik Habbinga, Kris Strecker, Walter Wong, Joshua Root,
                 Fabrice Bacchella, Zhenghua Xue, Qin Li, Darren Sawyer,
                 Vangel Bojaxhi, Ben England, Vikentsi Lapa.

    Run began: Thu Dec  6 21:44:04 2018

    Include fsync in write timing
    O_DIRECT feature enabled
    Auto Mode
    File size set to 10240 kB
    Record Size 4 kB
    Record Size 16 kB
    Record Size 512 kB
    Record Size 1024 kB
    Record Size 16384 kB
    Command line used: iozone -e -I -a -s 10M -r 4k -r 16k -r 512k -r 1024k -r 16384k -i 0 -i 1 -i 2
    Output is in kBytes/sec
    Time Resolution = 0.000001 seconds.
    Processor cache size set to 1024 kBytes.
    Processor cache line size set to 32 bytes.
    File stride size set to 17 * record size.
                                                              random    random     bkwd    record    stride                                    
              kB  reclen    write  rewrite    read    reread    read     write     read   rewrite      read   fwrite frewrite    fread  freread
           10240       4     2661     2977    13919    14392    10982     4107                                                          
           10240      16     6004     7628    34826    34952    35106    17878                                                          
           10240     512    33570    32039    61211    62413    61537    31775                                                          
           10240    1024    30877    20863    64869    66056    65049    20667                                                          

iozone test complete.

And

Code:

$ dmesg | grep mmc
...
[    2.061270] mmc_host mmc0: Bus speed (slot 0) = 148500000Hz (slot req 150000000Hz, actual 148500000HZ div = 0)
[    2.328584] dwmmc_rockchip fe320000.dwmmc: Successfully tuned phase to 226
[    2.329224] mmc0: new ultra high speed SDR104 SDXC card at address 0001
[    2.330533] mmcblk0: mmc0:0001 ED4QT 119 GiB
[    2.336271]  mmcblk0: p1 p2 p3 p4 p5 p6 p7

[z4v4l]

Thank you very much! Nice to see these numbers, it's good that your SD card is this cool, apparently not only UHS-I works, but SDR104 does! Well, with ~148MHz, but that is another question and it ultimately goes to Rockchip. Thanks again.