Hi, On 2013-11-24 05:28, Santosh Shilimkar wrote:
On Saturday 23 November 2013 11:10 PM, Greg Kroah-Hartman wrote: > On Sat, Nov 23, 2013 at 10:45:37PM -0500, Santosh Shilimkar wrote: >> On Saturday 23 November 2013 10:44 PM, Greg Kroah-Hartman wrote: >>> On Sat, Nov 23, 2013 at 10:37:01PM -0500, Santosh Shilimkar wrote: >>>> @@ -250,8 +249,8 @@ int __init dma_contiguous_reserve_area(phys_addr_t size, phys_addr_t base, >>>> *res_cma = cma; >>>> cma_area_count++; >>>> >>>> - pr_info("CMA: reserved %ld MiB at %08lx\n", (unsigned long)size / SZ_1M, >>>> - (unsigned long)base); >>>> + pr_info("CMA: reserved %ld MiB at %pa\n", (unsigned long)size / SZ_1M, >>>> + &base); >>> >>> Why is this pr_info() at all? That's just noise, please move it to >>> pr_debug(). >>> >> Marek can comment better but I think its useful print to know CMA >> reserved memory size. > > Useful to who? > Useful to anyone wants to know the CMA usage on a platform. CMA size is configurable and platforms tend to use different sizes based on needs. The info don't appear in /proc/meminfo, so probably dmsg grep is easy enough to know how much CMA memory being used on a platform. That was my point.I don't have strong argument here against not making it pr_debug but was waiting for Marek's opinion on it.
If possible I would like to keep pr_info. It already helped me a lot while analyzing someone's logs to find why memory allocation failed. A simple search for "CMA" messages enabled me to quickly check if CMA has been enabled and configured properly or not. It also gives a curious user some information about the memory configuration (especially if he don't need CMA, he will investigate why kernel has reserved so much memory). Best regards -- Marek Szyprowski Samsung R&D Institute Poland -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
