On Jun 11, 2010, at 11:08 AM, Pyun YongHyeon wrote: > On Fri, Jun 11, 2010 at 11:06:06AM -0700, Marcel Moolenaar wrote: >> >> On Jun 11, 2010, at 10:50 AM, Pyun YongHyeon wrote: >> >>>>>> >>>>>> I'm not clear why you even need bounce buffers for RX. The chip >>>>>> supports 64bit addresses with no boundary or alignment restrictions. >>>>>> >>>>> >>>>> Some controllers have 4G boundary bug so bge(4) restricts dma >>>>> address space. >>>> >>>> That limitation should be reflected in the boundary attribute of the tag, >>>> not the lowaddr/highaddr attributes. >>>> >>> >>> Yes, but that needed more code. And I don't have these buggy >>> controllers so I chose more simple way that would work even though >>> it may be inefficient. >> >> Do you happen to know if one or both of the hardware I have access to >> is the "buggy" hardware? >> > > Yes, both devices below can not handle 4GB boundary crossing in DMA > state machine.
Thanks. I'll keep that in mind. If I have a few cycles I'll patch the kernel to allow 64-bit DMA addresses with a 4G boundary restriction and run that through stress2. -- Marcel Moolenaar xcl...@mac.com _______________________________________________ svn-src-all@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-all To unsubscribe, send any mail to "svn-src-all-unsubscr...@freebsd.org"
