Mark Johnson wrote: > The only way is to alloc the 16K contig chunks individually. Darn. I was hoping there was some magic somewhere that I didn't know about.
.. the other way would be to not to worry about letting the system split it up and get one big contiguous chunk Thanx, carlos > > > Carlos Cumming wrote: >> I'm trying to figure out if it is possible using the ddi_dma_xxx >> routines along w/ an appropriate ddi_dma_attr_t to get memory where >> the dmacookie.dmac_size's will all be modulo 16k (0x4000). >> >> What I'm doing is I need a bunch (256) of 16k (i.e., > one page) >> physically contiguous chunks of memory. Since I don't need 256 * 16k >> to be contiguous, I'd rather give the system the option of breaking >> up my memory request. However, each 16k chunk needs to be contiguous >> b/c of hardware constraints. >> >> Is there a way to do it? > > > > > > MRJ > _______________________________________________ driver-discuss mailing list driver-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/driver-discuss
