Issue 120312

Summary:	MetricField class can generate INVALID_FLOAT_EXCEPTION on some computers
Product:	General	Reporter:	Yuri Dario <ydario>
Component:	code	Assignee:	AOO issues mailing list <issues>
Status:	CLOSED FIXED	QA Contact:
Severity:	Normal
Priority:	P3
Version:	4.0.0-dev
Target Milestone:	4.0.0
Hardware:	PC
OS:	OS/2
Issue Type:	DEFECT	Latest Confirmation in:	---
Developer Difficulty:	---

Description Yuri Dario 2012-07-18 21:36:52 UTC

On some computers, both intel and amd cpu, the MetricField class can generate an invalid floating point operation while converting double to long.
See revision 1363127 for more info, but is not enough to fix OS/2 bug.

Until I can find a PC with this bug to properly analyze this bug (sending debug dlls to Japan is not real time debugging...), I have find a reasonable workaround to code.

Comment 1 Yuri Dario 2012-07-18 21:38:32 UTC

vcl: workaround to invalid float operation.
Sending        control/field.cxx
Committed revision 1363129.

Comment 2 hdu@apache.org 2013-07-11 11:56:39 UTC

Updated status to FIXED and target to release that will contain the fix.