[MCOL-4509] Get rid __float128 Created: 2021-01-21  Updated: 2023-07-01

Status: Open
Project: MariaDB ColumnStore
Component/s: None
Affects Version/s: None
Fix Version/s: Icebox

Type: Task Priority: Major
Reporter: Roman Assignee: Unassigned
Resolution: Unresolved Votes: 0
Labels: None

Issue Links:
PartOf
is part of MCOL-4820 Umbrella for tasks related to the dec... Open
Epic Link: ColumnStore Compatibility Improvements

 Description   

__float128 is used in the code to conduct different scale wide-DECIMAL values math.
However:

  • this datatype reduces the precision of the result to some degree b/c of the 15 bits spent for storing order of the float value

 Comments   
Comment by Roman [ 2021-01-21 ]

There is another interesting way to solve the compilation issue at AArch64, namely port AArch64 aware __float128 and parts of libquadmath from Kunpeng gcc. Here its src presumably.

Generated at Thu Feb 08 02:50:53 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.