From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tony Lindgren Subject: Re: Re: [PATCH 1/2] OMAP2+: DMA: Workaround for invalid source position Date: Fri, 4 Nov 2011 13:08:26 -0700 Message-ID: <20111104200826.GE31337@atomide.com> References: <1320070817-9638-1-git-send-email-peter.ujfalusi@ti.com> <1320070817-9638-2-git-send-email-peter.ujfalusi@ti.com> <20111103212756.GZ31337@atomide.com> <2603119.amz7RjSE3J@barack> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mho-03-ewr.mailhop.org ([204.13.248.66]:51534 "EHLO mho-01-ewr.mailhop.org" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1750779Ab1KDUIi (ORCPT ); Fri, 4 Nov 2011 16:08:38 -0400 Content-Disposition: inline In-Reply-To: <2603119.amz7RjSE3J@barack> Sender: linux-omap-owner@vger.kernel.org List-Id: linux-omap@vger.kernel.org To: =?utf-8?B?UMOpdGVy?= Ujfalusi Cc: linux-arm-kernel@lists.infradead.org, Jarkko Nikula , linux-omap@vger.kernel.org * P=C3=A9ter Ujfalusi [111104 02:00]: > On Thursday 03 November 2011 14:27:56 Tony Lindgren wrote: > >=20 > > Should these tests be done only after the errata re-read for both > > src and dst patches? Otherwise the errata will not be handled? >=20 > Yes that might be a good idea. > I was trying to locate the original errata description for DMA_ERRATA= _3_3, but=20 > it does not exist, or at least I can not find it. This only been ment= ioned in=20 > the kernel's comment. Seems it got added with commit d3c9be2f. > I'm not sure when this DMA_ERRATA_3_3 would have been in force. My gu= ess would=20 > be that if someone wants to read the src/dst position right after the= channel=20 > is disabled, but what would we expect it to return when the channel h= as been=20 > disabled? No idea.. Or read it after stopping it? > I mean what is the reasonable src/dst for a disabled channel? 0 is as= good as=20 > any other number, probably the programmed start of the DMA transfer w= ould be=20 > my best bet. > I think the errata description for DMA_ERRATA_3_3 is not correct, and= it is in=20 > fact to handle the case I'm also handling: Before the first DMA reque= st the=20 > CDAC is 0 (since we configured it to be), the CSAC contains _somethin= g_ (most=20 > of the time 0, but can be random number). > In some situation re-reading the src/dst position will give enough ti= me to=20 > receive the first DMA request, which will update CDAC/CSAC registers. Yeah could be since the handling is just an immediate re-read of the re= gister.. =20 > What do you think? Probably the safest thing to do is move your check after the re-read te= st in case there's more to the ERRATA_3_3. Regards, Tony -- To unsubscribe from this list: send the line "unsubscribe linux-omap" i= n the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html