Unable to push Part 2 with Part 1 no RST version

[Jeerachee]

I'm having trouble getting the chip to mount to be able to copy over the Part 2 UF2 file. I understand that I'm supposed to hold the VOL+ button for 20 seconds, but when exactly? I've tried holding it for a couple minutes both at boot and after injecting the Samd21_Update payload to no avail. Part 1 seems to be working properly otherwise as it's automatically putting the console into RCM mode but without Part 2 it's pretty much useless.

[mattyams]

I am guessing that your chip is already installed? You'll need to be able to press the actual button on your chip (assuming it has one) twice after entering SAMD Update mode, then plug in your computer USB. The folder should pop up on your computer then.

[Jeerachee]

I am guessing that your chip is already installed? You'll need to be able to press the actual button on your chip (assuming it has one) twice after entering SAMD Update mode, then plug in your computer USB. The folder should pop up on your computer then.

I may be mistaken, but I believe the no reset version of part 1 disables this feature

[mattytrog]

Oh sheet... I haven't pushed part 2. I dropped it in but I didn't do the final push. Best thing... Downgrade to other part 1 version. I could have sworn I pushed v9.1. it supports new chips etc etc. I'm in Scarborough on holiday. I'm guessing you cannot enter bootloader mode? Or is it just not accepting any uf2?

On Sun, 29 May 2022, 20:44 mattyams, @.***> wrote:

I am guessing that your chip is already installed? You'll need to be able to press the actual button on your chip (assuming it has one) twice after entering SAMD Update mode, then plug in your computer USB. The folder should pop up on your computer then.

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1140512697, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZHJK6NKS3HKKLOSGEDVMPCJZANCNFSM5XIRPBOQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>

[Jeerachee]

Yeah I can't enter bootloader mode, even by double pressing reset button

[mattytrog]

Ok... What version part 2 are u using?

On Sun, 29 May 2022, 21:02 mattytrog mattytrog, @.***> wrote:

The no reset version gets rid of the reset button because we use a register to check on reboot. Basically, with the new version, you hold vol+ for 20 seconds and it enters bootloader mode.

On Sun, 29 May 2022, 20:56 Jeerachee, @.***> wrote:

I am guessing that your chip is already installed? You'll need to be able to press the actual button on your chip (assuming it has one) twice after entering SAMD Update mode, then plug in your computer USB. The folder should pop up on your computer then. I may be mistaken, but I believe the no reset version of part 1 disables this feature

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1140514736, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZFT5JC7JCFZ4EYQYA3VMPDV3ANCNFSM5XIRPBOQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>

[Jeerachee]

Well I was going to use the one that was in Old/PART_2_LATEST, I'm not 100% sure what the old version on there is. Is there some way to check since I can't hold VOL+ to get into the menu anymore?

[mattytrog]

If you have flashed part 1, your chip should be "erased" and go straight to the uf2 bootloader. Led on trinket should be green. Part 1 triggers RCM. Nothing more. Part 2 actually boots the unit.

So if you hold reset switch(which still resets btw) then press pwr to power on, it should boot normal. If you are plugged into pc, a drive should pop up.

The chip is either blank(triggers RCM only, then goes straight back to bootloader mode), or you have part 2 on there, so you should be able to get to menu. Let me know.

On Sun, 29 May 2022, 21:32 mattytrog mattytrog, @.***> wrote:

Ok... Let's run through this...

If you have flashed a part 1 with RST disabled, that should have erased the chip. Thus, if you manage to boot into "normal" mode, a FUSEE drive should pop up.

You should then be able to replace or update your file...

If you cannot boot "normal" mode because Chip RCM is constantly triggering, power switch off, push back cover in(so reset switch is pressed, power on)... Switch should boot normally. And if you plug into pc, you will see a FUSEE drive.

On Sun, 29 May 2022, 21:18 Jeerachee, @.***> wrote:

Well I was going to use the one that was in Old/PART_2_LATEST, I'm not 100% sure what the old version on there is. Is there some way to check since I can't hold VOL+ to get into the menu anymore?

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1140518264, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZDN2WYW63QDSRBYVM3VMPGJJANCNFSM5XIRPBOQ . You are receiving this because you commented.Message ID: @.***>

[Jeerachee]

Okay, so holding the reset button while powering on the console does allow it to boot normally, but when plugging it into my computer, a drive does not show up.

[mattytrog]

The screen needs to be off. It won't show if screen is "awake".

On Sun, 29 May 2022, 21:49 Jeerachee, @.***> wrote:

Okay, so holding the reset button while powering on the console does allow it to boot normally, but when plugging it into my computer, a drive does not show up.

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1140522047, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZCJE6GNKHK3EFDYPSLVMPJ4FANCNFSM5XIRPBOQ . You are receiving this because you commented.Message ID: @.***>

[Jeerachee]

Alright, the screen is off with the switch plugged in, and no drive is coming up. When I first installed the no reset UF2 the drive showed up and then disconnected 3-4 times before not showing up again, which seemed a bit odd to me, if that helps at all.

[mattytrog]

Hmmm... That is odd. What led lights are you seeing on the chip? Can you see? What chip are you using? A trinket? A part 1 flash should erase the chip. And it should trigger RCM and constantly reboot to "SWITCHBOOT" or FUSEE drive(red breathing led). I've temporarily removed the no reset bootloader until I can push the correct part 2 when I get home.

Tell me what the led is doing

On Sun, 29 May 2022, 22:09 Jeerachee, @.***> wrote:

Alright, the screen is off with the switch plugged in, and no drive is coming up. When I first installed the no reset UF2 the drive showed up and then disconnected 3-4 times before not showing up again, which seemed a bit odd to me, if that helps at all.

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1140524956, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZAYJPJXQSCMKBEO22TVMPMKFANCNFSM5XIRPBOQ . You are receiving this because you commented.Message ID: @.***>

[Jeerachee]

I'm using a trinket. The LED isn't even turning on, I had assumed that was intended with the no reset bootloader.

[mattytrog]

Hmmm... Does anything happen if you hold vol+ down for longer than 4 second? Any flashes?

If not, then your chip is actually looking for the uf2 and what you need to do, is plug in usb, power switch off... Then hold RST in and power on. Switch should boot. Release and just wait. The drive should pop open. I'm confident what's happening is the uf2 drive isn't being picked up by the pc.

On Sun, 29 May 2022, 22:27 Jeerachee, @.***> wrote:

I'm using a trinket. The LED isn't even turning on, I had assumed that was intended with the no reset bootloader.

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1140527599, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZFVHCTUEXKHNNJH2DLVMPONDANCNFSM5XIRPBOQ . You are receiving this because you commented.Message ID: @.***>

[Jeerachee]

When should I hold down VOL+?

[mattytrog]

At any time. If you have a part 2 flashed (which you shouldn't have if you have flashed part 1) then holding Vol+ down should start the chip options. You should see a flash approx every two seconds.

ChipRCM is being triggered as you need to hold down reset to get it to boot normally.

So... Part 1 is working- RCM is being triggered... There is no part 2 flashed. No LEDs. That means your chip is waiting for a pc connection.

You are hitting RCM every time. That means bootloader is working. You just need to get the pc to see it. Normally this is when the LCD turns off when the switch has fully booted. Or push Hekate and see if it pops up then.

On Sun, 29 May 2022, 22:42 Jeerachee, @.***> wrote:

When should I hold down VOL+?

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1140529863, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZBVAKFXP55MZMAY2TDVMPQD3ANCNFSM5XIRPBOQ . You are receiving this because you commented.Message ID: @.***>

[Jeerachee]

I just checked the continuity of my wires and they seem good and solid, holding VOL+ after booting normally, while screen is off but plugged in didn't do anything, no blinking LED. I tried what you just suggested about powering it on and waiting and it didn't show up. I tried on two other computers. It's reliably booting into RCM when I don't hold reset, no AutoRCM enabled through hekate. I'll try pushing hekate and see if the drive shows up from there. If it thinks there's a part 2 flashed for some reason, is there any way for me to force the bootloader to present the drive?

EDIT: Plugging the switch in after pushing hekate doesn't seem to do anything, but perhaps that's because the screen doesn't turn off?

[mattytrog]

If all else fails, and this becomes a real problem, I'll pull the release altogether and I'll write an automatic recovery routine. In the meantime, let's get your unit sorted.

The single colour LED(if fitted) should be breathing. The old part 1 used to have both red single LED and RGB enabled on a trinket by default. But there are new power on init routines in the new version.

I'm happy to either: Sort your console. Or Send me your chip. I'll Reflash your part 1 and part 2 on my jig which requires no computer Or I'll send you a replacement if you send me your non functional one.

This has been tested and retested on the newer qtpy chip as well as the others. I just didn't push part 2 before I came away.

I'm a bit limited what I can do as I'm not home. But if it is bricked, it is easily sorted. At no cost to you.

And I'll revise the release.

On Sun, 29 May 2022, 23:40 mattytrog mattytrog, @.***> wrote:

Part 2 relies on part 1. Never the other way round.

Flashing part 1 normally erases the whole chip. Or it should do. Once the chip has started(the first thing it does is drop the straps to gnd to trigger RCM), the pointer moves to application start(part 2). If this isn't present, or cannot run, it will stay in bootloader mode. Ie constantly showing the uf2 drive. You are getting no LEDs at all... Which would normally lead me to think that part 1 has failed. But you are getting into RCM every time. So it hasn't failed.

It's as though part 1 is running and it's jumping to a corrupted part 2. But you would at least get some kind of LED. Even if it's stuck on. I'm guessing you have removed the red led from the trinket(just using the RGB)

On Sun, 29 May 2022, 23:27 Jeerachee, @.***> wrote:

I just checked the continuity of my wires and they seem good and solid, holding VOL+ after booting normally, while screen is off but plugged in didn't do anything, no blinking LED. I tried what you just suggested about powering it on and waiting and it didn't show up. I tried on two other computers. It's reliably booting into RCM when I don't hold reset, no AutoRCM enabled through hekate. I'll try pushing hekate and see if the drive shows up from there. If it thinks there's a part 2 flashed for some reason, is there any way for me to force the bootloader to present the drive?

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1140536318, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZEBQBQZH5F25EVSUHTVMPVOXANCNFSM5XIRPBOQ . You are receiving this because you commented.Message ID: @.***>

[Jeerachee]

Is it able to be reflashed with a jig without having to desolder from the console? If so I wouldn't mind paying for a jig myself to do it. I use the console regularly and would hate to have to desolder and resolder it

[mattytrog]

You need to be able to get to the swd pads on the bottom. It's how we flash virgin chips.

On Mon, 30 May 2022, 00:41 Jeerachee, @.***> wrote:

Is it able to be reflashed with a jig without having to desolder from the console? If so I wouldn't mind paying for a jig myself to do it. I use the console regularly and would hate to have to desolder and resolder it

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1140547880, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZEE6H5FTQF762QSTRTVMP6C3ANCNFSM5XIRPBOQ . You are receiving this because you commented.Message ID: @.***>

[Jeerachee]

So it's going to have to be desoldered either way? There's no way I can push the payload that Part 2 includes manually in order to enter the bootloader drive through the chip assistant menu like the readme mentions?

[mattytrog]

It's looking like that. If there is no other way of getting the uf2 drive to show.... But part 1 is running... And part 2 shouldn't even exist because you flashed part 1. It's very peculiar. There will be a new version of the no reset part 1 where this cannot happen again.

If it was a bad part 1 flash, I could understand it. But RCM is being triggered. So it isn't a bad flash.

The Fusee drive should be showing. In fact you shouldn't be able to get rid of it until a part 2 is flashed.

I'll investigate when I get home. The new version will have a fail-safe in it. I didn't think it was necessary, on account f the chip being erased. So our options remain that I'll get a chip sent to you, or I'll fix it at no cost.

On Tue, 31 May 2022, 03:25 Jeerachee, @.***> wrote:

So it's going to have to be desoldered either way? There's no way I can push the payload that Part 2 includes manually in order to enter the bootloader drive through the chip assistant menu like the readme mentions?

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1141603360, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZD5DTAYY66HDCNADJ3VMV2CTANCNFSM5XIRPBOQ . You are receiving this because you commented.Message ID: @.***>

[Jeerachee]

If I get a jig set up, would I be able to dump the chip binary and send it to you to help with debugging? If so, I'd like to try fixing it myself if possible as it seems like a good learning experience, plus you're on holiday so I don't want to be a bother and I might be able to get it done before you're back

[mattytrog]

Ok... Well all the "jig" is, is a raspberry pi (with Ethernet for putty purposes), plug the poorly board into the usb and there are two swd pads on the bottom. I have an image on the gbatemp thread somewhere. It uses openocd. Can take a few attempts. Or remove your chip(keep wires attached to board) and fire it over to me. I know what's happened I think. For whatever reason, it's attempting to boot a "part 2". Let me explain the differences between the RST and non RST version... A delay of 500 mS and a strap hold time of 100 mS, Vs a delay of 0 and strap hold time of 75mS.

I made it for speed. And because (normally) you can now access the bootloader without the reset switch, I omitted the code for it. On the understanding that the chip would indeed wipe itself when a new part 1 was flashed. Every single chip I've tried(even the new ones, eg qtpy and Pi Pico) have operated in exactly the same way. But hearing how this has failed in your case, plus I didn't push the new version(it's sat there... New diagrams for new chips too etc), I've pulled the no RST version. People can only install an old version for the time being(I say old .. it's still current).

But yes that's the drill. Flash my image to a pi(pi zero/2/ maybe 3, not 4), plug into ethernet, connect through putty and connect/flash

On Tue, 31 May 2022, 10:11 Jeerachee, @.***> wrote:

If I get a jig set up, would I be able to dump the chip binary and send it to you to help with debugging? If so, I'd like to try fixing it myself if possible as it seems like a good learning experience, plus you're on holiday so I don't want to be a bother and I might be able to get it done before you're back

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1141877269, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZHZSOQBO3U6OMTENQLVMXJSLANCNFSM5XIRPBOQ . You are receiving this because you commented.Message ID: @.***>

[Jeerachee]

Well with the unfortunate unavailability of Raspberry PIs at the moment and any commonly recommended JTAG/SWD debuggers, I think I'll have to put fixing and sending you the currently flashed image on hold. It seems the fastest and most available course of action is to just get a new trinket, which still seem readily available for quite cheap.

[mattytrog]

Yep. Or a QTPY. Where about are you? You in UK?

On Tue, 31 May 2022, 20:13 Jeerachee, @.***> wrote:

Well with the unfortunate unavailability of Raspberry PIs at the moment and any commonly recommended JTAG/SWD debuggers, I think I'll have to put fixing and sending you the currently flashed image on hold. It seems the fastest and most available course of action is to just get a new trinket, which still seem readily available for quite cheap.

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1142546332, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZEVQRIE7NL3IIXFEODVMZQF3ANCNFSM5XIRPBOQ . You are receiving this because you commented.Message ID: @.***>

[Jeerachee]

I'm in the US

[mattytrog]

Ok... There is actually nothing wrong with your trinket. Just a bad flash. 100%fixable. Two more days I'll be home and can debug it properly. Though I know what's happened. So... Options are Reflash a basic bootloader using a raspberry pi jig, or get a fresh one. The old packages are just fine to use until I can get v9.1.1 out on Friday

On Wed, 1 Jun 2022, 17:45 Jeerachee, @.***> wrote:

I'm in the US

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1143863335, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZBAXK6AZ2Z7KVZL2Z3VM6HRZANCNFSM5XIRPBOQ . You are receiving this because you commented.Message ID: @.***>

[mattytrog]

Ok... Let's run through this...

If you have flashed a part 1 with RST disabled, that should have erased the chip. Thus, if you manage to boot into "normal" mode, a FUSEE drive should pop up.

You should then be able to replace or update your file...

If you cannot boot "normal" mode because Chip RCM is constantly triggering, power switch off, push back cover in(so reset switch is pressed, power on)... Switch should boot normally. And if you plug into pc, you will see a FUSEE drive.

On Sun, 29 May 2022, 21:18 Jeerachee, @.***> wrote:

Well I was going to use the one that was in Old/PART_2_LATEST, I'm not 100% sure what the old version on there is. Is there some way to check since I can't hold VOL+ to get into the menu anymore?

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1140518264, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZDN2WYW63QDSRBYVM3VMPGJJANCNFSM5XIRPBOQ . You are receiving this because you commented.Message ID: @.***>

[mattytrog]

The no reset version gets rid of the reset button because we use a register to check on reboot. Basically, with the new version, you hold vol+ for 20 seconds and it enters bootloader mode.

On Sun, 29 May 2022, 20:56 Jeerachee, @.***> wrote:

I am guessing that your chip is already installed? You'll need to be able to press the actual button on your chip (assuming it has one) twice after entering SAMD Update mode, then plug in your computer USB. The folder should pop up on your computer then. I may be mistaken, but I believe the no reset version of part 1 disables this feature

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1140514736, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZFT5JC7JCFZ4EYQYA3VMPDV3ANCNFSM5XIRPBOQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>

[mattytrog]

Part 2 relies on part 1. Never the other way round.

Flashing part 1 normally erases the whole chip. Or it should do. Once the chip has started(the first thing it does is drop the straps to gnd to trigger RCM), the pointer moves to application start(part 2). If this isn't present, or cannot run, it will stay in bootloader mode. Ie constantly showing the uf2 drive. You are getting no LEDs at all... Which would normally lead me to think that part 1 has failed. But you are getting into RCM every time. So it hasn't failed.

It's as though part 1 is running and it's jumping to a corrupted part 2. But you would at least get some kind of LED. Even if it's stuck on. I'm guessing you have removed the red led from the trinket(just using the RGB)

On Sun, 29 May 2022, 23:27 Jeerachee, @.***> wrote:

I just checked the continuity of my wires and they seem good and solid, holding VOL+ after booting normally, while screen is off but plugged in didn't do anything, no blinking LED. I tried what you just suggested about powering it on and waiting and it didn't show up. I tried on two other computers. It's reliably booting into RCM when I don't hold reset, no AutoRCM enabled through hekate. I'll try pushing hekate and see if the drive shows up from there. If it thinks there's a part 2 flashed for some reason, is there any way for me to force the bootloader to present the drive?

— Reply to this email directly, view it on GitHub https://github.com/mattytrog/FUSEE_SUITE/issues/2#issuecomment-1140536318, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ26VZEBQBQZH5F25EVSUHTVMPVOXANCNFSM5XIRPBOQ . You are receiving this because you commented.Message ID: @.***>