My M.2 SSD and adapter have arrived, and I've been investigating with
the tools described in the 'blog post that Paul linked to'
(https://lemariva.com/blog/2020/08/raspberry-pi-4-ssd-booting-enabled-trim).

I first of all investigated my existing Sandisk SATA SSD drive that's
connected via a cheapo SATA-to-USB adapter.  The indications are that
it's unmap-capable (but not enabled), and that it's using 'Bulk-Only
Transport' (BOT) and not UASP.  No great surprises there, and I guess I
could implement unmap as per the instructions in the blog post.


Code:
--------------------
    
  tc@pCPServer:/mnt/sdb2/tce$ fstrim -v /mnt/Music
  fstrim: FITRIM: Operation not permitted
  
  tc@pCPServer:/mnt/sdb2/tce$ sg_vpd -p bl /dev/sda
  .
  .
  .
  Maximum unmap LBA count: 4194240
  Maximum unmap block descriptor count: 1
  Optimal unmap granularity: 1 blocks
  Unmap granularity alignment valid: false
  Unmap granularity alignment: 0 [invalid]
  .
  .
  .
  
  
  tc@pCPServer:/mnt/sdb2/tce$ lsusb -v
  
  Bus 002 Device 003: ID 174c:55aa ASMedia Technology Inc. ASM1051E SATA 6Gb/s 
bridge, ASM1053E SATA 6Gb/s bridge, ASM1153 SATA 3Gb/s bridge, ASM1153E SATA 
6Gb/s bridge
  Device Descriptor:
  .
  .
  .
  Interface Descriptor:
  .
  . 
  bInterfaceClass         8 Mass Storage
  bInterfaceSubClass      6 SCSI
  bInterfaceProtocol     80 Bulk-Only
  .
  .
  .
  
--------------------


I then used the same commands on the new 'M.2 SATA drive'
(https://www.amazon.co.uk/gp/product/B078WYS5K6) connected via the Argon
adapter.  If I'm reading the output correctly, then this setup is NOT
unmap-capable, and is also using the BOT driver, rather than UASP. 
That's a bit of a surprise, given the blurb on the 'Pi Hut page'
(https://thepihut.com/collections/raspberry-pi-cases/products/argon-one-m-2-raspberry-pi-4-case)
about UASP support.  Perhaps I need to do something to enable it, or
perhaps I'm misreading the output.


Code:
--------------------
    tc@pCP:~$ fstrim -v /mnt/m2
  fstrim: FITRIM: Operation not permitted
  
  tc@pCP:~$ sg_vpd -p bl /dev/sda
  .
  .
  .
  Maximum unmap LBA count: 0 [Unmap command not implemented]
  Maximum unmap block descriptor count: 0 [Unmap command not implemented]
  Optimal unmap granularity: 0 blocks [not reported]
  Unmap granularity alignment valid: false
  Unmap granularity alignment: 0 [invalid]
  .
  .
  .
  
  tc@pCP:~$ lsusb -v
  
  Bus 002 Device 002: ID 174c:55aa ASMedia Technology Inc. ASM1051E SATA 6Gb/s 
bridge, ASM1053E SATA 6Gb/s bridge, ASM1153 SATA 3Gb/s bridge, ASM1153E SATA 
6Gb/s bridge
  Device Descriptor:
  .
  .
  .
  Interface Descriptor:
  .
  . 
  bInterfaceClass         8 Mass Storage
  bInterfaceSubClass      6 SCSI
  bInterfaceProtocol     80 Bulk-Only
  .
  .
  .
  
--------------------


I also did a few rudimentary speed tests, as per the instructions on
'this page' (https://elinux.org/RPi_SD_cards#Terminal_Commands).

I found that my old SATA drive achieves a write speed of around 150MB/s
and read speed of around 210MB/s.
The new M.2 SATA drive achieves a write speed of around 245MB/s and a
read speed of around 300MB/s.

These are all somewhat short of the 5Gbps possible under USB3, but hey,
at least the new M.2 drive seems to be about 50% faster than the SSD
I've been using.  And since it's only used to store a largely static
music collection, both drives are more than fast enough.


------------------------------------------------------------------------
chill's Profile: http://forums.slimdevices.com/member.php?userid=10839
View this thread: http://forums.slimdevices.com/showthread.php?t=113575

_______________________________________________
unix mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/unix

Reply via email to