From mboxrd@z Thu Jan 1 00:00:00 1970 From: peter.pearse@arm.com (Peter Pearse) Date: Mon, 9 Aug 2010 10:11:02 +0100 Subject: [PATCH 2/2] DMAENGINE: driver for the ARM PL080/PL081 PrimeCells In-Reply-To: References: <1277811391-17966-1-git-send-email-linus.walleij@stericsson.com> <20100805071235.GA436@morgana.i.gnudd.com> Message-ID: <000001cb37a2$ca081cb0$9644010a@Emea.Arm.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Linus Barriers probably just me being ultra cautious Given you've tested on multicore, feel free to remove them Peter > -----Original Message----- > From: Linus Walleij [mailto:linus.ml.walleij at gmail.com] > Sent: 06 August 2010 23:55 > To: Dan Williams > Cc: Alessandro Rubini; Peter Pearse; viresh.kumar at st.com; > linux-arm-kernel at lists.infradead.org; > linux-kernel at vger.kernel.org; yuanyabin1978 at sina.com; > ben-linux at fluff.org; kgene.kim at samsung.com; rmk at arm.linux.org.uk > Subject: Re: [PATCH 2/2] DMAENGINE: driver for the ARM > PL080/PL081 PrimeCells > > 2010/8/6 Dan Williams : > > >?One > > thing that should be on the todo list is to review the > usage of memory > >barriers. ?I don't think they are serving the purpose that > you think > >they are. > > All of those are from Peters original implementation, I never > use any barriers so they actually looked alien to me. Unless > I get a clear rationale from Peter I'll just remove them all > for the time being in the fix-up version due after updating > to generic slave control. > > Yours, > Linus Walleij >