Re: [PATCH crypto 2/2] crypto: safexcel - Map AEAD buffers with accurate DMA directions

From: Ralf Lici

Date: Fri Nov 13 2026 - 12:20:23 EST


Hi Herbert,

A gentle ping on this. Does the tested failure with the fully
DMA_BIDIRECTIONAL mapping address your concern, or do you still see a
problem with the per-entry DMA direction approach?

If the approach looks reasonable, I can respin the series with the
physical/logical layout and the A/B result described explicitly in the
commit message.

Thanks!

--
Ralf Lici
Mandelbit Srl