Merge in sp24 drivers/platforms #54
Conversation
[Common] Makefile Helpers
…ingup Fall 2024" This reverts commit 4e42c87.
…ingup Fall 2024" This reverts commit 4e42c87.
WIP: [BearlyML'24] Add DMA and quantized transformer drivers
Merge in individual driver code
|
@Fi50 @doihead @jasmangle REVIEW ME PLZ :3 |
|
Reviewed, looks mostly good to me. There should be RCC drivers included with this as well for Bearly24 and DSP24 to replace the current clock selection struct and allow for tile reset control, but I can copy those over fairly quickly (will need to test this tomorrow, but I'll be in the lab for a bit so I should be able to). The clock select struct was made hastily as a quick fix approach to getting a simple PLL startup and shouldn't be coupled with the Intel driver, since the clock selection ultimately happens through PRCI. Will merge ucb-bar#5 when I've tested it. |
| } | ||
|
|
||
| void pwm_set_frequency(PWM_Type *PWMx, uint32_t idx, uint32_t freq) { | ||
| // TODO: implementation |
There was a problem hiding this comment.
Should we maybe remove all the TODO: implementation except where we actually need it else becomes confusing
There was a problem hiding this comment.
Would be nice to add a header comment regarding the extent PWM was tested but is kind of a cherry on top >_<
| reg_write8(DMA_BASE, 0); | ||
| } | ||
|
|
||
| void enable_Crack(){ |
There was a problem hiding this comment.
If we're gonna upstream this.......... should we change it >_>
|
Yeah I didn't analyze too deeply, but left some small suggestions, I think once the PR is ready we should also pass it by @T-K-233 before merging into ucb-bar. I don't see us changing anything that could break a flow for anyone tho. |
|
Bump (maybe around now is a good time to get the merges sorted out?) |
6 participants
No description provided.