On Thu, 2017-06-22 at 21:00 +0530, Vinod Koul wrote: > On Thu, Jun 22, 2017 at 04:50:57PM +0200, Stefan Roese wrote: > > On 22.06.2017 15:00, Vinod Koul wrote: > > > On Tue, Jun 20, 2017 at 02:20:49PM +0200, Stefan Roese wrote: > > > > On 14.06.2017 10:25, Vinod Koul wrote: > > > > > On Wed, May 24, 2017 at 09:00:48AM +0200, Stefan Roese wrote: > > > > > > > > > > > +#define MSGDMA_MAX_TRANS_LEN 0xffffffff > > > > > > > > > > GENAMSK? > > > > > > > > I'm personally not a big fan of GENMASK. BIT(x) is good for > > > > single > > > > bits, but GENMASK usually looks lees clear - at least to me. So > > > > if > > > > you do not insist in this change, I would like to keep the > > > > masks. > > > > > > Well GENMASK(14, 11) tell me that it means mask of bit 14 thru 11, > > > which is > > > way clear than 0x7800 :) > > > > In this specific case, I see your point and would be willing to make > > this change gladly. But in the case above, its a the largest number > > of a u32 variable. I find it more confusing using GENMASK here. > > Probably there is a macro already for this u32(-1) value, that I > > should use instead. > > I think there is something, but can't recall which one. Andy? Since it's a maximum length of transfer on the one hand and in the hardware is actually a mask (bits are used for transfer length) you may choose one of U32_MAX GENMASK(31, 0) My personal preference is GENMASK(), since it's slightly closer to what hardware does inside. In case I'm wrong I would like to hear hardware engineer speaking. -- Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx> Intel Finland Oy -- To unsubscribe from this list: send the line "unsubscribe dmaengine" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
