Messages in this thread | | | Subject | Re: [PATCH 8/9] x86: ramdisk info print with high bits. | From | "H. Peter Anvin" <> | Date | Fri, 16 Nov 2012 12:08:46 -0800 |
| |
Use %#010llx
Yinghai Lu <yinghai@kernel.org> wrote:
>On Fri, Nov 16, 2012 at 11:39 AM, H. Peter Anvin <hpa@zytor.com> wrote: >> On 11/16/2012 11:21 AM, Yinghai Lu wrote: >>>> >>>> NAK, this is expected to match the resource print format (%pR), >which >>>> prints 10 digits by default and then expands. Furthermore, >printing >>>> *18* digits is downright silly since we still don't have 72-bit >addressing. >>> >>> that is the same as in e820_print_map:: >>> >>> printk(KERN_INFO "%s: [mem %#018Lx-%#018Lx] ", who, >>> (unsigned long long) e820.map[i].addr, >>> (unsigned long long) >>> (e820.map[i].addr + e820.map[i].size - 1)); >>> >>> >>> that is for 64bit address. >>> that extra 2 is for "0x" >>> >> >> Oh, right. It's the use of %#... I usually use 0x%... so I didn't >think >> of it. >> >>> or you prefer to cast them to pointer and use %pR for them all? >>> >>> or fix printk to add extra 2 for "0x" when # is found? >> >> We should normally use %pR or the equivalent format. The only reason >we >> do it different for e820_print_map is because it prints a whole bunch >of >> lines in which we want the columns to line up. > >so we just use #llx or #Lx instead in this patch?
-- Sent from my mobile phone. Please excuse brevity and lack of formatting.
| |