From mboxrd@z Thu Jan 1 00:00:00 1970 From: khilman@deeprootsystems.com (Kevin Hilman) Date: Wed, 10 Nov 2010 09:56:46 -0800 Subject: [PATCH v3 10/13] OMAP: DMA: Convert DMA library into DMA platform Driver In-Reply-To: (Manjunath Kondaiah G.'s message of "Wed, 10 Nov 2010 22:53:16 +0530") References: <1288099513-1854-1-git-send-email-manjugk@ti.com> <1288099513-1854-11-git-send-email-manjugk@ti.com> <8739rakubj.fsf@deeprootsystems.com> <87sjz9cfkv.fsf@deeprootsystems.com> Message-ID: <878w11awpt.fsf@deeprootsystems.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org "G, Manjunath Kondaiah" writes: >> >> > +#define OMAP_DMA4_LOGICAL_DMA_CH_COUNT 32 /* >> >> REVISIT: Is this 32 + 2? */ >> >> > + >> >> >> >> There are no users of this in this patch, all users are removed. >> > >> > This macro is used in hwmod data base. Do you mean hard coding this >> > value in hwmod db and remove header? >> >> What I mean is that from a reviewers perspective of this patch, you >> added a new #define that is not used anywhere in this patch. >> a #define >> should be added along with any users of it. > > No. This macro is used in omap2+ hwmod db through plat/dma.h which includes > mach/dma.h Which is not part of this patch. Again, as a general rule, don't introduce code (#defines, headers, functions, anything) in a patch if it is not used. >> >> That being said, the number of channels of the DMA IP sounds like >> something that should indeed be associated with the hwmod as it is a >> fixed HW feature. >> > If so, I can remove this macro and hard code in hwmod db. OK, good. Kevin