[Bug 741045] New: audiorate can can lose timestamp precision in some cases

[GStreamer (bugzilla.gnome.org)]

https://bugzilla.gnome.org/show_bug.cgi?id=741045
  GStreamer | gst-plugins-base | git

           Summary: audiorate can can lose timestamp precision in some
                    cases
    Classification: Platform
           Product: GStreamer
           Version: git
        OS/Version: Linux
            Status: UNCONFIRMED
          Severity: normal
          Priority: Normal
         Component: gst-plugins-base
        AssignedTo: gstreamer-bugs at lists.freedesktop.org
        ReportedBy: crh184 at psu.edu
         QAContact: gstreamer-bugs at lists.freedesktop.org
     GNOME version: ---


Created an attachment (id=292025)
 View: https://bugzilla.gnome.org/attachment.cgi?id=292025
 Review: https://bugzilla.gnome.org/review?bug=741045&attachment=292025

The git diff of the proposed change

Description:

The audiorate element uses gst_util_uint64_scale_int() for some time stamp
calculations, which truncates (rounds down), instead of
gst_util_uint64_scale_int_round(), which will round up when appropriate.  Over
time very small time stamp drifts can occur in some circumstances.  For code
that requires extremely strict time stamp consistency, this is a problem.  The
attached diff simply updates these functions.

Steps to Reproduce:

I confess that the case where I encounter this is in a pipeline with custom
elements (doing gravitational-wave data analysis, which demands very precise
time stamp bookeeping).  If this change is controversial, I will put the time
into coming up with a test case using "stock" elements.  Please let me know.

Actual Results:

Timestamps can drift by approximately 1 ns over time

Expected results:

Exact timestamps that reflect the sample offset and the sample rate

Build Date & Platform:

patch against git checkout on Ubuntu 14.04. First noticed on 0.10.32 on
Scientific Linux 6.

The git diff is attached.

-- 
Configure bugmail: https://bugzilla.gnome.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.
You are the assignee for the bug.