Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Update loops in CpuMath to be more efficient #1177

Merged
merged 4 commits into from
Oct 31, 2018
Merged

Update loops in CpuMath to be more efficient #1177

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
828b9e0
Update loops to be more efficient
jwood803 Oct 6, 2018
1893294
Make temp variable for loop
jwood803 Oct 15, 2018
2b03984
Update SSE case and where loop variable gets declared
jwood803 Oct 28, 2018
a100d66
Remove extra new line
jwood803 Oct 29, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 5 additions & 3 deletions src/Microsoft.ML.CpuMath/AvxIntrinsics.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -22,6 +22,8 @@ internal static class AvxIntrinsics

private const int Vector256Alignment = 32;

private const int destinationEnd = pDstEnd - 4;
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Where is pDstEnd defined, I wouldn't expect this could be a constant....


[MethodImplAttribute(MethodImplOptions.AggressiveInlining)]
private static bool HasCompatibleAlignment(AlignedArray alignedArray)
{
Expand Down Expand Up @@ -431,7 +433,7 @@ public static unsafe void AddScalarU(float scalar, Span<float> dst)

Vector128<float> scalarVector128 = Sse.SetAllVector128(scalar);

if (pDstCurrent + 4 <= pDstEnd)
if (pDstCurrent <= destinationEnd)
{
Vector128<float> dstVector = Sse.LoadVector128(pDstCurrent);
dstVector = Sse.Add(dstVector, scalarVector128);
Expand Down Expand Up @@ -517,7 +519,7 @@ public static unsafe void ScaleSrcU(float scale, Span<float> src, Span<float> ds

Vector128<float> scaleVector128 = Sse.SetAllVector128(scale);

if (pDstCurrent + 4 <= pDstEnd)
if (pDstCurrent <= destinationEnd)
{
Vector128<float> srcVector = Sse.LoadVector128(pSrcCurrent);
srcVector = Sse.Multiply(srcVector, scaleVector128);
Expand Down Expand Up @@ -563,7 +565,7 @@ public static unsafe void ScaleAddU(float a, float b, Span<float> dst)
Vector128<float> a128 = Sse.SetAllVector128(a);
Vector128<float> b128 = Sse.SetAllVector128(b);

if (pDstCurrent + 4 <= pDstEnd)
if (pDstCurrent <= destinationEnd)
{
Vector128<float> dstVector = Sse.LoadVector128(pDstCurrent);
dstVector = Sse.Add(dstVector, b128);
Expand Down
6 changes: 3 additions & 3 deletions src/Microsoft.ML.CpuMath/SseIntrinsics.cs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -417,7 +417,7 @@ public static unsafe void AddScalarU(float scalar, Span<float> dst)

Vector128<float> scalarVector = Sse.SetAllVector128(scalar);

while (pDstCurrent + 4 <= pDstEnd)
while (pDstCurrent <= pDstEnd - 4)
Copy link
Member

@tannergooding tannergooding Oct 24, 2018 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Could you update the SSE case as well?

{
Vector128<float> dstVector = Sse.LoadVector128(pDstCurrent);
dstVector = Sse.Add(dstVector, scalarVector);
Expand Down Expand Up @@ -479,7 +479,7 @@ public static unsafe void ScaleSrcU(float scale, Span<float> src, Span<float> ds

Vector128<float> scaleVector = Sse.SetAllVector128(scale);

while (pDstCurrent + 4 <= pDstEnd)
while (pDstCurrent <= pDstEnd - 4)
{
Vector128<float> srcVector = Sse.LoadVector128(pSrcCurrent);
srcVector = Sse.Multiply(srcVector, scaleVector);
Expand Down Expand Up @@ -512,7 +512,7 @@ public static unsafe void ScaleAddU(float a, float b, Span<float> dst)
Vector128<float> aVector = Sse.SetAllVector128(a);
Vector128<float> bVector = Sse.SetAllVector128(b);

while (pDstCurrent + 4 <= pDstEnd)
while (pDstCurrent <= pDstEnd - 4)
{
Vector128<float> dstVector = Sse.LoadVector128(pDstCurrent);
dstVector = Sse.Add(dstVector, bVector);
Expand Down