Hi, On Fri, Jan 15, 2016 at 7:40 AM, Jaehoon Chung <jh80.chung@xxxxxxxxx> wrote: > Hi, Alim. > > On 01/15/2016 10:20 PM, Alim Akhtar wrote: >> Hi Jaehoon, >> >> On Mon, Jan 11, 2016 at 3:15 PM, Jaehoon Chung <jh80.chung@xxxxxxxxxxx> wrote: >>> It's fixed "SDMMC_##" as prefix into mci_read/write. >>> So it's difficult to debug or read which offset is used. >>> If it's passed by original defined register name, it can be checked >>> which offset is used, and more readability than now. >>> >> >> From commit message it is not clear what problem/difficulty you faced >> with the current setup? >> and how does this patch help to resolve the same? >> Can you put some more light on the same? > > Thanks for comments. :) > Indeed, there is no problem..It's just difficult to search for the defined location with "ctags". > > Even though we know that it's related with CTRL register, > it's difficult to find where CTRL is defined, which offset is used. > I used the 'grep', 'cscope' command or other methods. > If using this patch, we can search the defined location more easier than now with "ctags". > > When dwmmc controller was implemented, it was defined with "SDMMC_##" in mci_read/write() function. > But i didn't find the benefit about wrapping the register name. > > Actually, it should be my preference. > If you agree my preference, i will resend the patch with commit msg in more detail. I agree with your preference. It's a bunch of churn that touches lots of lines in the file, but it definitely helps when searching through the code, especially for those not familiar with it. -Doug -- To unsubscribe from this list: send the line "unsubscribe linux-mmc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
