feat(U-Boot): Add User Guide on Key Writer Lite #294
Conversation
Harsha-VM
requested review from
cshilwant,
StaticRocket,
praneethbajjuri,
uditkumarti,
aniket-l and
VeeruPrudhvi
as code owners
github-actions
bot
assigned bryanbrattlof, devarsht, jmenti, praneethbajjuri, r-vignesh and uditkumarti
github-actions
bot
requested review from
bryanbrattlof,
devarsht,
jmenti and
r-vignesh
source/linux/Foundational_Components/U-Boot/UG-General-Info.rst
Outdated
Show resolved
Hide resolved
source/linux/Foundational_Components/U-Boot/UG-Key-Writer-Lite.rst
Outdated
Show resolved
Hide resolved
source/linux/Foundational_Components/U-Boot/UG-Key-Writer-Lite.rst
Outdated
Show resolved
Hide resolved
source/linux/Foundational_Components/U-Boot/UG-Key-Writer-Lite.rst
Outdated
Show resolved
Hide resolved
source/linux/Foundational_Components/U-Boot/UG-Key-Writer-Lite.rst
Outdated
Show resolved
Hide resolved
| In the case of TI AM62L PROC181E1 EVMs, an I2C driver is necessary to send command packets to the IO expander, | ||
| which then toggles the IO pin connected to the Vpp pin, thereby controlling the power supply to the pin. |
There was a problem hiding this comment.
Will this procedure change with the other EVM revisions?
There was a problem hiding this comment.
The procedure as such in general will remain same like one has to select the i2c bus to which the chip that powers up Vpp pin is connected to, and then turn off Vpp, configure Vpp port as output and turn on Vpp.
But, The values of the commands can change if the i2c schematic changes and if the IO expander chip's address that enables Vpp changes in other EVM revisions.
The SoC user can design a suitable circuit that enables the Vpp pin to be powered. This is an example for the user in the above mentioned EVM revision.
|
|
||
| Changes made to efuses, by programming them, take effect (like become | ||
| visible in Memory-Mapped Registers (MMRs), device type change etc.) | ||
| after a complete System-on-Chip (SoC) power cycle. |
There was a problem hiding this comment.
Do we have any error conditions? What do we do if something in this chain doesn't go according to plan?
There was a problem hiding this comment.
The error conditions of fuse writebuff (the corressponding TISCI keywriter lite API it calls) is documented in the TISCI docs.
| u-boot=> dcache flush | ||
| u-boot=> md 0x82000000 | ||
|
|
||
| #. Efuse modification requires a voltage to be applied on a specific pin (Vpp) during the programming. |
There was a problem hiding this comment.
I'm wondering if we should re-order this. Should setting up the EVM come first, then creating the structures, then sending the structures to TIFS? IDK I've never had to convert an FS chip to SE
There was a problem hiding this comment.
I feel this flow is fine, where you setup Vpp right before calling fuse writebuff
Harsha-VM
force-pushed
the
key-writer-lite-docs
branch
from
11a39b9 to
9aba367
Compare
Add a user guide that introduces and gives an example flow of key writer lite from u-boot. Signed-off-by: Harsha Vardhan V M <[email protected]>
Harsha-VM
force-pushed
the
key-writer-lite-docs
branch
from
9aba367 to
01e7204
Compare
|
@bryanbrattlof @StaticRocket to re-review |
Add a user guide that introduces and gives an example flow of key writer lite from u-boot.