Struct DbufRead 

pub struct DbufRead<'a, 'b, T: Copy, const N: usize> { /* private fields */ }

Continuous DMA-read typestate handle for double-buffer mode.

The DMA engine reads M0 and M1 alternately in continuous transfer. The CPU is the producer and may fill the inactive bank through a BankGuard<Read, _, _> obtained from DbufRead::bank_guard.

Implementations

impl<'a, 'b, T: Copy, const N: usize> DbufRead<'a, 'b, T, N>

pub fn m0_dma_addr(&self) -> DmaAddr

DMA bus address of the M0 bank (stable for the typestate’s lifetime).

pub fn m1_dma_addr(&self) -> DmaAddr

DMA bus address of the M1 bank.

pub fn bank_guard<'g, C: DcaCache>( &'g mut self, _ctx: &mut DcaCacheCtx<'_, C>, current_target: Bank, ) -> BankGuard<'g, Read, T, N>

Acquire a guard over the inactive bank.

current_target names the bank the engine is currently servicing (read from the stream’s CT bit). The guard exposes the opposite bank.

No cache op is emitted at construction. Per DCB-00 §5 (DCB-01b-A 2026-05-03 amendment), the Read direction’s cache op fires at release — the CPU is about to write new data via the guard’s slice, so cleaning now would publish stale data. The clean at release publishes the just-completed CPU writes before the engine flips back to this bank.

_ctx is unused at construction for Read direction (kept for symmetry with DbufWrite::bank_guard, which uses it for the entry-side invalidate).

pub fn stop_double_buffer<C: DcaCache>( self, ctx: &mut DcaCacheCtx<'_, C>, ) -> DbufCpu<'a, 'b, T, N>

Stop the double-buffer transfer and transition back to DbufCpu.

Caller MUST stop the engine before calling. Cache op: DcaCache::clean over both banks.