Re: zero filling a storage device (was: dd and mbr)

From: Ralf Mardorf <ralf-mardorf_at_riseup.net>
Date: Fri, 14 Jan 2022 18:13:03 UTC
On Fri, 14 Jan 2022 17:14:18 +0000, Arthur Chance wrote:
>On 14/01/2022 15:32, Christian Weisgerber wrote:
>> "Kevin P. Neal":
>>   
>>> Are we certain that an SSD won't at least track that there is
>>> nothing written to a logical block and therefore it must be all
>>> zeros? I'm not 100% that an SSD will always keep a logical block
>>> assigned to a physical block. And I'm not 100% certain that an SSD
>>> won't notice that all zeros are being written to a block and just
>>> optimize out the write.  
>> 
>> It's tempting to speculate that an SSD could treat an all-zeros
>> block write effectively like a TRIM.
>> 
>> I'll note that there are SSDs that compress the data written to
>> them.  (Compression in storage devices isn't new.  Terry Welch's
>> 1984 paper, where he presented the LZW compression algorithm, already
>> talks about this.)
>>   
>
>May I suggest "man trim"? (From 12.1 onwards.)

*?*

Nothing on the operating system side of the SSD's controller (and its
firmware) has got direct access to what's under the hood of the SSD.
Due to wear leveling a SSD much likely keeps a lot of not really erased
sensitive data. This data is not accessible/restoreable without
replacing the firmware by a forensic tool or by an unhappy coincidence,
but replacing firmware is way more likely, than a high-tech lab to
restore secret data and even an unhappy coincidence isn't far too
unlikely.