FreeCalypso > hg > freecalypso-tools
annotate doc/How-flash-really-works @ 1000:39a6090a052a
doc/How-flash-really-works: article written
author | Mychaela Falconia <falcon@freecalypso.org> |
---|---|
date | Sat, 09 Dec 2023 09:08:19 +0000 |
parents | |
children |
rev | line source |
---|---|
1000
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
1 How NOR flash memory really works |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
2 ================================= |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
3 |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
4 The type of flash memory used in Calypso GSM devices is formally known as NOR |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
5 flash. Most embedded software programmers and tinkerers know the fundamental |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
6 principle of how NOR flash works: any bit can be transitioned from a '1' to a |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
7 '0' at any time in any combination (an operation called programming), but the |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
8 opposite transition (from '0' to '1' bits, an operation called erasure) can only |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
9 be done on fairly large sectors - you can erase a sector and make it all 1s, |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
10 but changing bits from 0 to 1 individually or in any smaller granularity |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
11 (smaller than a sector) is impossible. |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
12 |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
13 What many "software-minded" programmers and tinkerers don't realize, however, |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
14 is that sector erasure is not an elementary or atomic operation that magically |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
15 "makes all bits 1s" in one motion. Instead it is a complex process with two or |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
16 three substeps: |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
17 |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
18 1) Before starting the physical process of erasure, one has to go through all |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
19 bits in the to-be-erased sector and make them all 0s. Any bits that are in |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
20 '1' state when the sector erase operation is commanded MUST be programmed to |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
21 '0' state before the actual erasure begins! In the language of flash chip |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
22 industry, this step is called preprogramming. In the case of flash chips |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
23 that are used in Calypso GSM devices (all known ones), this preprogramming |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
24 step is done internally by the chip, so that you as the user or software |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
25 developer are not aware of it - but it is there nonetheless. The chip does |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
26 NOT magically "wave" all bits in the sector into '1' state, instead it first |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
27 makes them all '0' internally, and only then erases. |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
28 |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
29 2) Once every bit in the sector is in '0' state, the real physics of erasure |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
30 begins. All bit cells in the sector are physically acted upon at once in |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
31 this step, and because it is a probabilistic process involving a Gaussian |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
32 distribution, all bit cells need to be in the fully programmed state before |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
33 they begin their shared journey toward the erased state. |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
34 |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
35 3) The step of preprogramming every bit to 0 prior to erasure prevents the |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
36 absolutely unacceptable condition of gross overerasure - but given the |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
37 Gaussian distribution, some bit cells may still get a little overerased. |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
38 Many (most? all? not sure) flash chips therefore implement a third internal |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
39 step before the software-visible "erase" operation is declared complete: |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
40 they go through all bit cells in the just-erased sector, check for |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
41 overerasure, and "soft-program" (move slightly to the right in the Vt |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
42 distribution) any overerased cells. This step is called post-erase |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
43 conditioning or recovery. |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
44 |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
45 The above process was originally explained to me (Mother Mychaela) some years |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
46 ago (around 2008, IIRC) by a Spansion support engineer on a conference call at |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
47 my then day job - it was a project for a customer who was big and powerful |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
48 enough to get top-tier support from chip vendors. More recently, however, some |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
49 other flash vendors have posted public documents that provide the same |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
50 explanation - here is one from Renesas/Adesto: |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
51 |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
52 https://www.freecalypso.org/pub/embedded/flash/REN_an500_APN_20210702_1.pdf |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
53 |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
54 Even though the above document was written by Renesas (or more precisely, the |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
55 part that was originally Adesto), the theory described therein applies just as |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
56 well to Intel, Spansion and Samsung flash chips that are used in Calypso GSM |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
57 devices. For anyone who wishes to know how NOR flash memory really works, I |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
58 strongly recommend reading that Renesas appnote - it is a good description. |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
59 |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
60 Additional note on terminology: describing the two states of a flash memory cell |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
61 as '0' and '1', like I did above, is only a convenience for software-minded |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
62 people. A more proper view is to think in terms of a "programmed state" and an |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
63 "erased state" for each bit cell. History and tradition are such that flash |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
64 chips return '0' on read in the programmed state and '1' in the erased state |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
65 (this tradition probably originates from the fact that the actual NV storage |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
66 element, a transistor, conducts read current in the erased state), at least for |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
67 the main flash array - however, when flash memory elements are used for |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
68 additional purposes such as write protection controls, it is best to think |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
69 natively in terms of programmed and erased states. For the latter kind of |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
70 special applications, an opposite polarity may be applied in read-bit values. |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
71 |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
72 One straightforward take-away from this theory is that flash endurance is really |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
73 about program-erase cycles, rather than number of program or number of erase |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
74 operations. Every time you give a sector erase command, every bit in that |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
75 sector cycles through the fully programmed (0) state first before becoming |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
76 erased (1), irrespective of whether or not you programmed into it on your own! |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
77 Hence every bit-cell of the affected sector always goes through a full |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
78 program-erase cycle, and all bits in a given sector are always cycled equally, |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
79 irrespective of whether they get written with mostly-0s or mostly-1s in between |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
80 erase cycles. |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
81 |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
82 Another situation where this raw physics gets exposed to the user is the case |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
83 of special-purpose non-volatile bits in flash chips outside of the main flash |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
84 memory array - for example, Persistent Protection Bits (PPBs) in some Spansion |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
85 and Samsung flash chips. While program and erase commands for the main flash |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
86 array invoke chip-internal mechanisms that take care of everything and present |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
87 a sane model of 0s and 1s to software, Spansion PL-J PPB program and erase |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
88 commands expose raw guts: there is a command that applies a raw program pulse |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
89 to a single PPB, and there is a command that applies a raw erase pulse to the |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
90 NV memory element (like a little sector of its own) that holds all PPBs. |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
91 Applying the erase pulse without preprogramming every PPB first would be very |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
92 bad (see Renesas appnote about the badness of overerasure) - hence in a seeming |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
93 paradox, one has to explicitly lock every sector before applying PPB erase |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
94 pulses that will eventually unlock everything! |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
95 |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
96 Our flash ppb-erase-all command does implement the preprogramming step before |
39a6090a052a
doc/How-flash-really-works: article written
Mychaela Falconia <falcon@freecalypso.org>
parents:
diff
changeset
|
97 actual erasure, and the present document (hopefully) explains why. |