ESP8266 Developer Zone

Re: Bad flash writes with v1.5.1 (increased power usage???)

2017-07-19T07:15:02+08:00

I don't have a flash_write_flash function... I have an rboot-ota.c rboot_write_flash function - I'm still (SDK 2.1.0) getting occasional E:M 5600 or thereabouts messages which mess up the OTA which of course makes RBOOT's OTA code fail...

Any ideas? I can't see a fix incorporated anywhere.

Statistics: Posted by scargill — Wed Jul 19, 2017 7:15 am

Re: Bad flash writes with v1.5.1 (increased power usage???)

2016-04-09T17:22:03+08:00

Thank you, useful to know what the maximum could be.

Statistics: Posted by rab — Sat Apr 09, 2016 5:22 pm

Re: Bad flash writes with v1.5.1 (increased power usage???)

2016-04-07T15:16:26+08:00

Hi,

The TCP window is 4 x MSS = 4 x 1460 = 5840 > 4096 (Sector size).
So the received data length may be larger than the sector size.

Thanks for your interest in ESP8266 !

Statistics: Posted by ESP_Faye — Thu Apr 07, 2016 3:16 pm

Re: Bad flash writes with v1.5.1 (increased power usage???)

2016-04-06T21:28:30+08:00

Oh wow, that's a little embarrassing. Of course I was aware of that limitation, I wrote the comment in the code afterall, but I didn't think this situation could ever occur! I had never seen packets of data arrive longer than about mid 1400s previously. I assumed these were limited to the MSS, or something similar, which would always be much smaller than the 4k sector size. Since you suggested looking at this I've done some more detailed testing in this area and found an occasional packet of 5744 arrive with SDK v1.5.2. Is this a new increased limit? Or has something just changed the timing in the SDK which allows a few packets to build up occasionally? Interestingly 5744 is still the only other size I have seen, after my previous max of 1436 - no other values have occurred yet.

To test further I created interference at the RF level to try and affect the internal working of the receive and was successful in causing these large packets to occur more frequently (still all 5744 though, odd!). This allowed me to confirm the problem and the fix.

Thank you very much for your help. I'll update rBoot and get some users to test it for real. If you can shed some light on why it's just started happening after v1.5.1 and a bit more on the inner workings of the network stack that would be interesting.

Statistics: Posted by rab — Wed Apr 06, 2016 9:28 pm

Re: Bad flash writes with v1.5.1 (increased power usage???)

2016-04-06T18:30:38+08:00

Hi,

In your reboot-ota.c, flash_write_flash function

Code:

   if (len > SECTOR_SIZE) {
      // to support larger writes we would need to erase current
      // (if not already done), next and possibly later sectors too
   } else {
      // check if the sector the write finishes in has been erased yet,
      // this is fine as long as data len < sector size
      if (status->last_sector_erased != (status->start_addr + len) / SECTOR_SIZE) {
         status->last_sector_erased = (status->start_addr + len) / SECTOR_SIZE;
         spi_flash_erase_sector(status->last_sector_erased);
      }
   }

You need to deal with the situation that "len > SECTOR_SIZE", otherwise, it will cause the problem.

Statistics: Posted by ESP_Faye — Wed Apr 06, 2016 6:30 pm

Re: Bad flash writes with v1.5.1 (increased power usage???)

2016-04-06T18:12:07+08:00

Also having the same issue. With 50 devices worldwide that use OTA, this is a big problem for me.
Tony

Statistics: Posted by spants — Wed Apr 06, 2016 6:12 pm

Re: Bad flash writes with v1.5.1 (increased power usage???)

2016-04-02T17:11:47+08:00

Sorry, but the problem is not fixed. I still get intermittent bad flashes with the test application on github when using the new library.

Richard.

Statistics: Posted by rab — Sat Apr 02, 2016 5:11 pm

Re: Bad flash writes with v1.5.1 (increased power usage???)

2016-03-23T17:31:57+08:00

Hi,

Please have a try with the attachment based on ESP8266_NONOS_SDK_V1.5.2.

If your problem is still unsolved, please feel free to let us know.

libmain_for_ESP8266_NONOS_SDK_V1.5.2.zip

Statistics: Posted by ESP_Faye — Wed Mar 23, 2016 5:31 pm

Re: Bad flash writes with v1.5.1 (increased power usage???)

2016-03-23T09:12:38+08:00

Same problem here on both ESP-01 and ESP-12.

Statistics: Posted by giox — Wed Mar 23, 2016 9:12 am

Re: Bad flash writes with v1.5.1 (increased power usage???)

2016-03-08T05:29:52+08:00

I can reproduce this with Richard's test app. I notice that in all failed cases the "unwritten" bytes (the 0xff bytes) start at the start of a sector. Given the way the test app works I'm wondering if spi_flash_erase_sector is returning before completion and actually being applied after the spi_flash_write which immediately follows it (and straddles sectors), although this sounds rather odd.

I wrote my own simpler test app to see if I can repro it with that. I can't so far. The key different being my test app isn't doing anything active with the network, so not drawing as much power.

Statistics: Posted by piersfinlayson — Tue Mar 08, 2016 5:29 am

Re: Bad flash writes with v1.5.1 (increased power usage???)

2016-03-05T20:02:22+08:00

Any update on this bug? It's causing a real problem for users and preventing upgrade to latest SDK.

Statistics: Posted by rab — Sat Mar 05, 2016 8:02 pm

Re: Bad flash writes with v1.5.1 (increased power usage???)

2016-02-14T04:53:06+08:00

This is of great concern for us as we are using rBoot and are now left in a situation where we can´t upgrade to any of the newer SDK´s since OTA fails every time for us using the newer SDK´s.

//Henrik Pedersen

Statistics: Posted by henkep — Sun Feb 14, 2016 4:53 am

Re: Bad flash writes with v1.5.1 (increased power usage???)

2016-02-14T04:09:01+08:00

p.s. examples of bad writes are included on github so you can see what has happened.

Statistics: Posted by rab — Sun Feb 14, 2016 4:09 am

Re: Bad flash writes with v1.5.1+ (increased power usage???)

2016-02-14T04:07:10+08:00

I have created a (fairly simple) test case:
https://github.com/raburton/flashwritebug
Instruction are in the readme.txt
I have tested the code with the new v1.5.2 sdk and the bug is still there. The flash write gets blocks of 0xff bytes (I assume these are erased, but not rewritten) in the middle of the file. This seems to happen on any sustained write, so I assume it will affect the sdk bootloader ota write just as much as as rBoot ota.

Richard.

Statistics: Posted by rab — Sun Feb 14, 2016 4:07 am

Re: Bad flash writes with v1.5.1 (increased power usage???)

2016-01-25T17:59:01+08:00

Hi,

Sorry that we could not duplicate your problem, could you provide your test code for debugging ?

Statistics: Posted by ESP_Faye — Mon Jan 25, 2016 5:59 pm

Bad flash writes with v1.5.1 (increased power usage???)

2016-01-22T00:39:51+08:00

I've found that flash writes sometimes fail (silently) with sdk v1.5.1, but they used to work fine with older sdks. Some users of rBoot have noticed this and I have been able to reproduce it myself just by recompiling with sdk v1.5.1. All the spi_flash_write calls return success, but after writing a rom (~250k) there may be one of more unwritten areas (0xff) in the flash. The number, position and size of the unwritten area varies each time.

I have seen this before when users have not had enough power - too few mA from their power supply - to flash and use wifi at the same time (and the write still does not return an error). Now it is happening on devices that were fine before sdk v1.5.1 so I suspect that the new sdk is using more power than the old version.

To make it worse if you try to read the flash back to check the data is correct, it appears to come from the cache and so it looks fine (but if you use esptool.py to read the flash you can see it is bad). This means it is not possible to check the write worked at all!

Can you advise on how to fix this?

Richard.

Statistics: Posted by rab — Fri Jan 22, 2016 12:39 am