That did it! Ill go ahead and print a case and get rockin with this thing.
Excellent!Statistics: Posted by mck1117 — Wed May 19, 2021 5:50 am
]]>
]]>
]]>
Statistics: Posted by AndreyB — Tue May 18, 2021 10:06 pm
]]>
]]>
Somehow I was able to get my not default v12 tune into both of these units you and David have so I am a bit puzzled on what's the defect but seems like we have one 
The first flash works - subsequent erase operations don't work. The first one looks like it worked because it was already erased, but the next erase didn't work. We were erasing the wrong page on 1MB devices. Part way fixed, will PR tomorrow (Monday) probably.Statistics: Posted by mck1117 — Mon May 17, 2021 6:06 am
]]>
How exactly does that work? so the "tune" portion is written to separate memory locations at the same time for redundancy, is that what im understanding?
The biggest thing is the fact that you cannot just update 100 bytes on stm32f7 flash - you have to erase and write complete memory page and those are not your SD card 512 bytes pages, those are you to 512K pages - i.e. to update 100 bytes you have to really update all half a megabyte, and if power is turned off you corrupt all content not just your 100 bytes.To make things more reliable yes we update two clones of tune in two of these pages. Thus we need to know position of those two for all chips we support. Because of chip shortage drama we had no choice but to fabricate 0.6 with a version of stm32 we have not used previously. It comes with minor differences we have not paged proper attention at. We did not handle this 0.6 to my own satisfaction, you should not be the one reporting issues that trivial
Somehow I was able to get my not default v12 tune into both of these units you and David have so I am a bit puzzled on what's the defect but seems like we have one Statistics: Posted by AndreyB — Mon May 17, 2021 6:04 am
]]>
How exactly does that work? so the "tune" portion is written to separate memory locations at the same time for redundancy, is that what im understanding? I just managed to get the code to compile on my machine early today cause i thought maybe there was a build target directly for this chip that i could compile but I'm a very limited in knowledge on some of this stuff and trying desperately to fill the gaps in. managed to compile it but couldnt find a way to flash the ecu with my own binary, I'm guessing i need to understand the openocd CLI right? such a learning curve I apologize in advance...
Statistics: Posted by biohazardav — Mon May 17, 2021 5:35 am
]]>
]]>
FWIW I'm getting the same error.
Neat. I'll give it a try on the F7 0.6 I have here.Statistics: Posted by mck1117 — Sun May 16, 2021 11:39 pm
]]>
]]>
Statistics: Posted by AndreyB — Sun May 16, 2021 10:15 pm
]]>
"QC special build" is failing to communicate "please do not touch me I am special quality control build that's not what real users should be using"
we should put non-consumer builds inside a folder called "special" or somethingStatistics: Posted by mck1117 — Sun May 16, 2021 9:53 pm
]]>
mck1117 this unit was switched to proper nDBANK settings since it has departed with a not failing non-QC snapshot build.
Statistics: Posted by AndreyB — Sun May 16, 2021 9:00 pm
]]>
@mck1117 ideas?
Statistics: Posted by AndreyB — Sun May 16, 2021 8:57 pm
]]>