Path: ...!eternal-september.org!feeder3.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: pozz <pozzugno@gmail.com>
Newsgroups: comp.arch.embedded
Subject: Re: Library for save an events log in Flash
Date: Thu, 18 Apr 2024 22:36:36 +0200
Organization: A noiseless patient Spider
Lines: 140
Message-ID: <uvs08l$2f2pb$1@dont-email.me>
References: <uvrb97$24nd4$1@dont-email.me> <uvrscc$2e2vi$1@dont-email.me>
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Injection-Date: Thu, 18 Apr 2024 22:36:38 +0200 (CEST)
Injection-Info: dont-email.me; posting-host="58db72a803fee8d7202d536cb26bb54a";
	logging-data="2591531"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX18ZZrQo/IAwCh9SW1TOGetCALxOqPZJM20="
User-Agent: Mozilla Thunderbird
Cancel-Lock: sha1:Of0fB3weaM7s0Iu9/e60UfZQMzA=
Content-Language: it
In-Reply-To: <uvrscc$2e2vi$1@dont-email.me>
Bytes: 7289

Il 18/04/2024 21:30, David Brown ha scritto:
> On 18/04/2024 16:38, pozz wrote:
>> The request is very common: when some interested events occur in the 
>> system, they, with the related timestamp, must be saved in a log. The 
>> log must be saved in an external SPI Flash connected to a MCU. The log 
>> has a maximum number of events. After the log is completely filled, 
>> the new event overwrite the oldest event.
>>
>> I tried to implement such library, but I eventually found it's not a 
>> simple task, mostly if you want a reliable library that works even 
>> when errors occur (for example, when one writing fails).
>>
>> I started by assigning 5 sectors of SPI Flash to the log. There are 
>> 256 events in a sector (the event is a fixed struct).
>> In this scenario, the maximum log size is 1024 events, because the 5th 
>> sector can be erased when the pointer reaches the end of a sector.
>>
>> The first challenge is how the system can understand what is the first 
>> (newest) event in the log at startup. I solved saving a 16-bits 
>> counter ID to each event. The initialization routine starts reading 
>> all the IDs and taks the greatest as the last event.
>> However initially the log is empty, so all the IDs are 0xFFFF, the 
>> maximum. One solution is to stop reading events when 0xFFFF is read 
>> and wrap-around ID at 0xFFFE and not 0xFFFF.
> 
> Start at 0xfffe, and count down.  


And what to do when the counter reaches zero? It can wrap-around up to 
0xfffe (that is very similar to an increasing counter from 0x0000-0xFFFE).

 > Or xor with 0xffff for storage.  Or
> wrap at 0xfffe, as you suggest.  Or use 32-bit values.  Or have another 
> way to indicate that the log entry is valid.  

I will add a CRC for each entry and that can be used to validate the 
event. An empty/erased slot filled with 0xFF will not pass CRC validation.


> Or, since you have a 
> timestamp, there's no need to track an ID - the timestamp will be 
> increasing monotonically.

I don't want to use timestamps for two reasons:

- the system wall clock can be changed (the system is isolated)
- the library I'm writing doesn't know the content of "events", for
   it the event is an opaque sequence of bytes.


>> However there's another problem. What happens after writing 65535 
>> events in the log? The ID restarts from 0, so the latest event hasn't 
>> the greatest ID anymore.
>>
>> These are the saved IDs after 65536 events:
>>
>>      1^ SECT    2^ SECT    3^ SECT    4^ SECT    5^SECT---------->
>>      0xFB00 ... 0xFC00 ... 0xFD00 ... 0xFE00 ... 0xFF00 ... 0xFFFF
>>
>> The rule "newest event has greatest ID" is correct yet. Now a new 
>> event is written:
>>
>>      1^ SECT-------> 2^ SECT   3^ SECT   4^ SECT   5^SECT--------->
>>      0x0000 0xFB01.. 0xFC00 .. 0xFD00 .. 0xFE00 .. 0xFF00 .. 0xFFFF
>>
>> Now the rule doesn't work anymore. The solution I found is to detect 
>> discontinuity. The IDs are consecutive, so the initialization routine 
>> continues reading while the ID(n+1)=ID(n)+1. When there's a gap, the 
>> init function stops and found the ID and position of the newest event.
> 
> Make your counts from 0 to 256*5 - 1, then wrap.  Log entry "n" will be 
> at address n * sizeof(log entry), with up to 256 log entries blank. Then 
> you don't need to store a number at all.

What do you mean with log entry "0"? Is it the oldest or the newest? I 
think the oldest, because that formula is imho correct in this case.

However it doesn't appear correct when the log has rotated, that happens 
after writing 5x256+1 events. In this case the newest entry ("n"=1024) 
is at address 0, not n*sizeof(entry).


>> But he problems don't stop here. What happens if an event write 
>> failed? Should I verify the real written values, by reading back them 
>> and comparing with original data? In a reliable system, yes, I should.
>>
>> I was thinking to protect an event with a CRC, so to understand at any 
>> time if an event is corrupted (bad written). However the possibility 
>> to have isolated corrupted events increase the complexity of the task 
>> a lot.
> 
> An 8-bit or 16-bit CRC is peanuts to calculate and check.  

I know. Here the increased complexity wasn't related to the CRC 
calculation, but to the possibility to have isolated corrupted slots in 
the buffer. Taking into account these corrupted slots isn't so simple 
for me.


> Write the 
> whole log entry except for a byte or two (whatever is the minimum write 
> size for the flashes), which must be something other than 0xff's.  Check 
> the entry after writing if you really want, then write the final byte or 
> two.  Then if there's a power-fail in the middle, it's obvious that the 
> log entry is bad.

I don't understand why to write in two steps. I think I can write all 
the bytes of an entry, including CRC. After writing, I can read back all 
the bytes and check integrity.


> For SPI NOR flash, the risk of a bad write - other than through 
> unexpected resets or power fails - is negligible for most purposes. 
> Don't over-complicate things worrying about something that will never 
> happen.

Indeed unexpected resets, but mainly power failes, are possible. Rare, 
but possible.


>> Suppose to write the 4th event with ID=3 at position 4 in the sector. 
>> Now suppose the write failed. I can try to re-write the same event at 
>> poisition 5. Should I use ID=4 or ID=5? At first, I think it's better 
>> to use ID=5. The CRC should detect that at position 4 is stored a 
>> corrupted event.
>> After that two events are written as well, ID=6 and 7.
>>
>> Now the higher application wants to read the 4th event of the log 
>> (starting from the newest). We know that the newest is ID=7, so the 
>> 4th event is ID=7-4+1=4. However the event with ID=4 is corrupted, so 
>> we should check the previous ID=3... and so on.
>> Now we can't have a mathematical function that returns the position of 
>> an event starting from it's ordinal number from the newest event.
>>
>>
>> Eventually I think it's better to use a public domain library, if it 
>> exists.
>>
>