From mboxrd@z Thu Jan 1 00:00:00 1970 From: tsbogend@alpha.franken.de (Thomas Bogendoerfer) Subject: Re: [PATCH] mips: Add dma_mmap_coherent() Date: Fri, 22 Aug 2008 16:36:22 +0200 Message-ID: <20080822143622.GA8413@alpha.franken.de> References: <1219255088.3258.45.camel@localhost.localdomain> <1219326912.3265.2.camel@localhost.localdomain> <20080821214118.GA12516@alpha.franken.de> <20080822094131.GA6717@alpha.franken.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: ralf@linux-mips.org, James Bottomley , linux-mips@linux-mips.org, Parisc List To: Takashi Iwai Return-path: In-Reply-To: List-ID: List-Id: linux-parisc.vger.kernel.org On Fri, Aug 22, 2008 at 12:23:48PM +0200, Takashi Iwai wrote: > unsigned long prot = pgprot_val(_prot) & ~_CACHE_MASK; > #ifdef CONFIG_SGI_IP32 > #ifdef CONFIG_CPU_R10000 > prot = prot | _CACHE_UNCACHED_ACCELERATED; > #else > prot = prot | _CACHE_CACHABLE_NO_WA; > #endif > #else > prot = prot | _CACHE_UNCACHED; > #endif > return __pgprot(prot); this won't work for recording channels on IP32, because the write trough mapping will hide updates done via DMA. I'd start with just prot |= _CACHE_UNCACHED and if some MIPS system needs more specific treatment, we just add that later. Thomas. -- Crap can work. Given enough thrust pigs will fly, but it's not necessary a good idea. [ RFC1925, 2.3 ]