f3probe false positive (partially)

[axet]

Hello!

I recently brought a flash drive which been reported as "counterfeit of type limbo" using f3probe command with 64GB out of 128GB.

Bus 001 Device 003: ID 048d:1234 Integrated Technology Express, Inc. Chipsbank CBM2199 Flash Drive

I also run my hddtest.py script (from previous issue) and full surface+crc test shows up that drive is only 16% badblocks and actual size is 100GB. Reading / writing speeds are also very bad.

Here are test logs:

F3 probe 8.0
Copyright (C) 2010 Digirati Internet LTDA.
This is free software; see the source for copying conditions.

WARNING: Probing normally takes from a few seconds to 15 minutes, but
         it can take longer. Please be patient.

Probe finished, recovering blocks... Done

Bad news: The device `/dev/sdb' is a counterfeit of type limbo

You can "fix" this device using the following command:
f3fix --last-sec=134217727 /dev/sdb

Device geometry:
	         *Usable* size: 64.00 GB (134217728 blocks)
	        Announced size: 117.19 GB (245760000 blocks)
	                Module: 128.00 GB (2^37 Bytes)
	Approximate cache size: 3.00 MB (6144 blocks), need-reset=no
	   Physical block size: 512.00 Byte (2^9 Bytes)

Probe time: 14'37"
 Operation: total time / count = avg time
      Read: 1.58s / 23305 = 68us
     Write: 14'36" / 55956 = 15.6ms
     Reset: 2us / 2 = 1us

writing test:
99% 245757983  7111606296 

duration:  11:42:01
block size:  512
total sectors:  245760000 (117.2GiB)
sectors processed:  245760000 100% (117.2GiB)
filter count:  1

per sector writing speed:
250ns 4362240 1%
500ns 25159680 10%
5000ns 206604288 84%
10000ns 661504 0%
100000ns 3674254 1%
1000000ns 48036 0%
5000000ns 1249989 0%
50000000ns 3952904 1%
200000000ns 32768 0%
600000000ns 14336 0%
3000000001ns 1 0%

writing speeds:
  5% 3.8MiB/s
 10% 3.3MiB/s
 15% 3.2MiB/s
 20% 3.0MiB/s
 25% 3.2MiB/s
 30% 3.1MiB/s
 35% 3.2MiB/s
 40% 2.0MiB/s
 45% 2.1MiB/s
 50% 1.1MiB/s
 55% 3.5MiB/s
 60% 3.6MiB/s
 65% 3.5MiB/s
 70% 3.4MiB/s
 75% 3.3MiB/s
 80% 3.7MiB/s
 85% 3.5MiB/s
 90% 3.2MiB/s
 95% 3.3MiB/s
100% 3.3MiB/s

reading test:
83% 204974080  70818296  error
...
99% 245759999  9726  error

duration:  2:09:09
block size:  512
total sectors:  245760000 (117.2GiB)
sectors processed:  245760000 100% (117.2GiB)
filter count:  40785920
badblocks count: 40785920 16% (19.4GiB)

per sector reading speed:
5000ns 85 0%
10000ns 468 0%
100000ns 245759439 99%
1000000ns 1 0%
5000000ns 1 0%
50000000ns 6 0%

reading speeds:
  5% 16.6MiB/s
 10% 15.8MiB/s
 15% 15.8MiB/s
 20% 15.8MiB/s
 25% 15.8MiB/s
 30% 15.8MiB/s
 35% 15.8MiB/s
 40% 15.8MiB/s
 45% 15.9MiB/s
 50% 15.8MiB/s
 55% 15.8MiB/s
 60% 16.0MiB/s
 65% 16.1MiB/s
 70% 16.1MiB/s
 75% 16.2MiB/s
 80% 16.0MiB/s
 85% 15.1MiB/s
 90% 13.8MiB/s
 95% 13.8MiB/s
100% 13.8MiB/s

[AltraMayor]

Hi @axet,

I only support the F3 tools. You are welcome to document your tools elsewhere and leave a link on this thread so others can find them. However, I do not have time to properly compare the output of f3probe with those of your tools.

[axet]

I'm not reporting here about my tools.

I'm saying f3probe false positive on a device. f3probe claims that flash drive I'm using is counterfeit limbo, which I assume, means it overwrite existing data. But that is not true. This flash drive simply has 20GB bad blocks at the end of drive (space which can be written but return corrupted data when read). The rest of the drive 100GB is fine. I guess algorithm which compute available space used in f3proble is wrong.

Same tests can be done using f3write / f3read.