Subject: Re: Machine-independent bus DMA interface proposal
To: None <abrown@eecs.harvard.edu, thorpej@nas.nasa.gov>
From: Chris Torek <torek@BSDI.COM>
List: tech-kern
Date: 09/30/1996 07:06:26
Actually, whether the IOMMU is separate is machine-dependent (IOMMU
PTEs are allowed to enter the regular TLB; see the description of
the TRCR's IL bit, on p. 89 of the microSPARC2 user's manual).
Nonetheless, it is still true that the sun4m gets hairy. :-)  The
IOMMU is logically separate even if it is physically on the same
chip.