Re: Issue awge(4) on NetBSD/CubieBoard2

[Martin Husemann <martin%duskware.de@localhost>]

On Sun, Dec 21, 2014 at 05:50:10PM +0900, MAEKAWA Masahide wrote:
> I found an issue of awge(4) on NetBSD/CubieBoard2.
> When TX ring is almost full, ``map'' (L858) has the valid DMA map,
> and overrided (L869) and destroyed (L870) by bus_dma(9).
> After this, tx_intr do bus_dmamap_sync(9) and panic.

I do not quite understand: in line 869 the map is not destroyed, but
just unloaded again. Also neither sc_txq.t_cur nor tx_queued have been
increased, so the sync range in tx_intr should not include the map we
failed to add and unloaded again.

Not saying there ain't no bug there, but I don't see how your description
of events would lead to a panic.

Martin