Bug 6957

Summary:	multsum_f64 is not optimized
Product:	liboil	Reporter:	Marcus Brubaker <aurelius.marcus>
Component:	unknown	Assignee:	David Schleef <ds>
Status:	REOPENED ---	QA Contact:	David Schleef <ds>
Severity:	enhancement
Priority:	high	CC:	aurelius.marcus, jdc+fd
Version:	HEAD
Hardware:	x86 (IA32)
OS:	Linux (All)
Whiteboard:
i915 platform:		i915 features:
Attachments:	multsum_f64_unroll8 and multsum_f64_sse2_unroll4 multsum.patch

Description Marcus Brubaker 2006-05-18 04:03:22 UTC

Multsum_f64 has only a reference implementation.  Patch with two optimized
versions to be attached.

Producing other levels of loop unrolling is easily doable.  Is it a good idea to
add these other versions?

Comment 1 Marcus Brubaker 2006-05-18 04:04:30 UTC

Created attachment 5660 [details] [review]
multsum_f64_unroll8 and multsum_f64_sse2_unroll4

Comment 2 David Schleef 2006-05-19 16:57:33 UTC

Applied.

Comment 3 Marcus Brubaker 2006-05-31 11:31:24 UTC

Created attachment 5770 [details] [review]
multsum.patch

This is a new version of the patch against latest anoncvs.  It does two things.
 First, it fixes the previously broken SSE2 implementation.  This version
actually speeds things up notably.  Witness: 

multsum_f64
  multsum_f64_unroll8
    ave=576 std=1.14286
  multsum_f64_sse2_unrollb
    ave=573 std=1.14286
  multsum_f64_sse2_unrolla
    ave=568 std=1.14286
  multsum_f64_sse2
    ave=576 std=1.14286
  multsum_f64_ref
    ave=850.444 std=1.16741

Second, it introduces an unstrided version of multsum for f32 and f64 with SSE2
optimized versions.  Results:

multsum_f32_ns
  multsum_f32_ns_sse
    ave=330 std=11.3928
  multsum_f32_ns_ref
    ave=737 std=1.125

multsum_f64_ns
  multsum_f64_ns_sse2_unroll2
    ave=372 std=1.14286
  multsum_f64_ns_sse2_unroll
    ave=382 std=1.14286
  multsum_f64_ns_sse2
    ave=463.444 std=3.59953
  multsum_f64_ns_ref
    ave=734 std=1.125

Comment 4 Marcus Brubaker 2006-05-31 11:33:35 UTC

Reopening because of new patch.

Comment 5 Marcus Brubaker 2006-06-07 14:06:34 UTC

There appears to be a bug in the unrolled SSE2 versions.  It doesn't manifest it
on my main dev machine (a Pentium M laptop) but shows up on another machine (a
Pentium 4).  I'm trying to track it down now.  Will report back when it's resolved.

Comment 6 David Schleef 2008-02-18 18:24:55 UTC

Patch doesn't apply.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.