CC1352P: CC1352P ECDSA_Sign issue

myh

Part Number: CC1352P

I have a issue about ECDSA_Sign,

I did adding ECDSACC26X2.c, h files and renaming sign to sign_mod thing, but it is not working.

So decided to open a new thread, old thread locked and not resolved. (That, TI Engineer Clément posted last message)

--> https://e2e.ti.com/support/wireless-connectivity/bluetooth/f/538/t/893329?tisearch=e2e-sitesearch&keymatch=ECDSA_sign#

over 6 years ago

0 myh over 6 years ago

Prodigy 205 points

As you can see, after i edited ti_drivers_config file (i know i should not edit that file, it is derived), i got an ECDSA_Sign return as "-1".
(#define ECDSA_STATUS_ERROR (-1) )

Still NOT working after that workaround.

I did change the included ECDSACC26X2 file that i included, not in <ti/drivers/...>

0 Furkan Fehmi Sefiloglu over 6 years ago in reply to myh

TI__Expert 3118 points

Hi Team,

This is customer request and we have critical project timeline.

Can you please support promptly.

Thanks in advance

Best Regards

Furkan Sefiloglu

0 myh over 6 years ago in reply to Furkan Fehmi Sefiloglu

Prodigy 205 points

Any progress?

0 Evan W over 6 years ago in reply to myh

TI__Mastermind 41670 points

Hi Myh,

Apologies for the extreme delay. We use a system where we filter posts that are unassigned. Unfortunately, because Furkan had responded to this thread, our system will automatically take the first post by a TIer and assign it to that person as it's assumed the team will have been the first person to respond. Because of this, it seems this post was not flagged as unassigned because it was mistakenly assigned to Furkan. I've asked someone on my team in the meantime to dig into this and provide feedback as soon as possible.

0 myh over 6 years ago in reply to Evan W

Prodigy 205 points

Thanks for reply, I tried nearly everything to solve that but no chance. Firstly i assumed, my curve is not supported (Curve25519), and change it nist256, but result is same. If anyone try and share me a solution that includes, - BLE connectable beacon advertise - TRNG, ECDH, ECDSA, SHA2, AES-GCM, activated.. It is really urgent, we believe that cc1352p is fit our product, and designed hardware, and produced that nearly 5K units now. But we cant solve cryptographic functions of our product with current situation.

0 myh over 6 years ago in reply to Evan W

Prodigy 205 points

Any progress about that?

0 Luis M over 6 years ago in reply to myh

TI__Expert 5345 points

myh,

I'm looking into your issue. I will provide an update on 7/15/2020.

Luis

0 Luis M over 6 years ago in reply to Luis M

TI__Expert 5345 points

myh,

I'm working with my team of experts to get you an answer and an example that can enable you to use our product.

This issue is a high priority for me. I should provide an update on 7/16/2020.

Thanks for your patience.

-Luis

0 myh over 6 years ago in reply to Luis M

Prodigy 205 points

Please also check your solution for Curve25519 too.

We have a strict deadline about that project, and that problem puts us 3 week behind the project timeline.

0 Luis M over 6 years ago in reply to myh

TI__Expert 5345 points

myh,

I understand your frustration. I'm going to do all I can to get a solution.

In fact, this is my update: I'm actively working with our RnD team to resolve this situation. They are aware of the name conflict issue. In the long term, they have created a ticket to provide a fix in a future SDK. In the short term, I myself am working with them to provide a clear and reliable workaround. Unfortunately, this will still take me a few more days. I'm targeting 7/23 to provide a solution. I'll post updates here.

Thanks,

Luis