[PATCHv2][Ping 1] Increase internal precision of ldbl-128ibm decimal printf

[Paul E. Murphy]

Thanks Joseph.  Committed as 37a4c70bd4c5c74ac562072e450dc02e8cb4c150.

On 03/30/2016 06:04 PM, Joseph Myers wrote:
> This patch is OK.
> 
> It's arguable how different this bug actually is from 5268 (the original 
> submission of bug 5268 doesn't clearly define the scope of the bug, and 
> the one-bit-loss issue here gets mentioned later in the bug 5268 
> comments).  But I suppose this bug can be considered more specific to the 
> one-bit-loss, with 5268 left for the more general case where there are 
> more than 106 bits of precision in the long double value.
>