Skip to content

Remove nan, inf and denorm #727

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

Closed
joaosaffran wants to merge 3 commits into from
Closed

Remove nan, inf and denorm #727

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
c5e1423
remove nan, inf and denorm
joaosaffran Feb 10, 2026
cf224eb
remove nan, inf, denorm from more tests
joaosaffran-zz Feb 10, 2026
8607269
modify log10 test
joaosaffran Feb 11, 2026
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
7 changes: 3 additions & 4 deletions test/Feature/HLSLLib/asin.16.test
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,17 +25,16 @@ Buffers:
- Name: In
Format: Float16
Stride: 8
Data: [ 0x7e00, 0xfc00, 0x8001, 0x8000, 0x0000, 0x0001, 0x7c00, 0x3c00, 0xbc00, 0x3e00, 0xbe00, 0x7e00,]
# NaN, -Inf, -denorm, -0, 0, denorm, Inf, 1, -1, 1.5, -1.5,
Data: [ 0x8000, 0x0000, 0x7c00, 0x3c00, 0xbc00, 0x3e00, 0xbe00]
# -0, 0, denorm, Inf, 1, -1, 1.5, -1.5,
- Name: Out
Format: Float16
Stride: 8
FillSize: 24
- Name: ExpectedOut # The result we expect
Format: Float16
Stride: 8
Data: [ 0x7e00, 0x7e00, 0x0000, 0x0000, 0x0000, 0x0000, 0x7e00, 0x3e48, 0xbe48, 0x7e00, 0x7e00, 0x7e00,]
# NaN, NaN, 0.0, 0.0, 0.0, 0.0, NaN, 1.570796, -1.570796, NaN, NaN,
Data: [ 0x0, 0x0, 0x7E00, 0x3E48, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0 ]
Results:
- Result: Test1
Rule: BufferFloatEpsilon
Expand Down
6 changes: 2 additions & 4 deletions test/Feature/HLSLLib/asin.32.test
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,17 +25,15 @@ Buffers:
- Name: In
Format: Float32
Stride: 16
Data: [ nan, -inf, -0x1.e7d42cp-127, -0, 0, 0x1.e7d42cp-127, inf, 1, -1, 1.5, -1.5, nan,]
# NaN, -Inf, -denorm, -0, 0, denorm, Inf, 1, -1, 1.5, -1.5,
Data: [ -0, 0, 1, -1, 1.5, -1.5,]
- Name: Out
Format: Float32
Stride: 16
FillSize: 48
- Name: ExpectedOut # The result we expect
Format: Float32
Stride: 16
Data: [ nan, nan, 0.0, 0.0, 0.0, 0.0, nan, 1.570796, -1.570796, nan, nan, nan,]
# NaN, NaN, 0.0, 0.0, 0.0, 0.0, NaN, 1.570796, -1.570796, NaN, NaN,
Data: [ 0, 0, 1.5708, -1.5708, 0, 0, 0, 0, 0, 0, 0, 0 ]
Results:
- Result: Test1
Rule: BufferFloatEpsilon
Expand Down
8 changes: 4 additions & 4 deletions test/Feature/HLSLLib/cosh.16.test
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,17 +25,17 @@ Buffers:
- Name: In
Format: Float16
Stride: 8
Data: [ 0x7e00, 0xfc00, 0x8001, 0x8000, 0x0000, 0x0001, 0x7c00, 0x3c00, 0xbc00, 0x7e00, 0x7e00, 0x7e00,]
# NaN, -Inf, -denorm, -0, 0, denorm, Inf, 1, -1,
Data: [0x8000, 0x0000, 0x3c00, 0xbc00, 0x8000, 0x0000, 0x3c00, 0xbc00, 0x8000, 0x0000, 0x3c00, 0xbc00]
# -0, 0, 1, -1, -0, 0, 1, -1, -0, 0, 1, -1,
- Name: Out
Format: Float16
Stride: 8
FillSize: 24
- Name: ExpectedOut # The result we expect
Format: Float16
Stride: 8
Data: [ 0x7e00, 0x7c00, 0x3c00, 0x3c00, 0x3c00, 0x3c00, 0x7c00, 0x3e2c, 0x3e2c, 0x7e00, 0x7e00, 0x7e00,]
# NaN, Inf, 1.0, 1.0, 1.0, 1.0, Inf, 1.543081, 1.543081,
Data: [0x3C00, 0x3C00, 0x3E2C, 0x3E2C, 0x3C00, 0x3C00, 0x3E2C, 0x3E2C, 0x3C00, 0x3C00, 0x3E2C, 0x3E2C]
# 1.0, 1.0, 1.543, 1.543, 1.0, 1.0, 1.543, 1.543, 1.0, 1.0, 1.543, 1.543
Results:
- Result: Test1
Rule: BufferFloatULP
Expand Down
6 changes: 2 additions & 4 deletions test/Feature/HLSLLib/frac.16.test
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,17 +25,15 @@ Buffers:
- Name: In
Format: Float16
Stride: 8
Data: [ 0x7e00, 0xfc00, 0x03FF, 0x8000, 0x0000, 0x7c00, 0xbc00, 0x4170, 0x63d1, 0xC764, 0x7e00, 0x7e00,]
# NaN, -Inf, 0.00006097555, -0, 0, Inf, -1, 2.719, 1000.5, -7.390625,
Copy link
Copy Markdown
Member

@farzonl farzonl Feb 10, 2026 •
edited
Loading

Choose a reason for hiding this comment

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

since the half is in hex we do need a comment to represent the numbers. Same below

Data: [ 0x03FF, 0x8000, 0x0000, 0xbc00, 0x4170, 0x63d1, 0xC764]
- Name: Out
Format: Float16
Stride: 8
FillSize: 24
- Name: ExpectedOut # The result we expect
Format: Float16
Stride: 8
Data: [ 0x7e00, 0x7e00, 0x03FF, 0x0000, 0x0000, 0x7e00, 0x0000, 0x39c1, 0x3800, 0x38E1, 0x7e00, 0x7e00,]
# NaN, NaN, 0.00006097555, 0, 0, NaN, 0, 0.719, 0.5, 0.6098633,
Data: [ 0x3FF, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0, 0x0 ]
Copy link
Copy Markdown
Member

@farzonl farzonl Feb 10, 2026

Choose a reason for hiding this comment

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

This test is failing now post your changes

OffloadTest-vk :: Feature/HLSLLib/frac.16.test

Results:
- Result: Test1
Rule: BufferFloatEpsilon
Expand Down
6 changes: 2 additions & 4 deletions test/Feature/HLSLLib/frac.32.test
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,17 +25,15 @@ Buffers:
- Name: In
Format: Float32
Stride: 16
Data: [ nan, -inf, 0, -0, 0, 1.5, inf, -1, 2.718280, 1000.599976, -7.389, nan,]
# NaN, -Inf, 0, -0, 0, 1.5, Inf, -1, 2.718280, 1000.599976, -7.389,
Data: [ 0, -0, 0, 1.5, -1, 2.718280, 1000.599976, -7.389, 0, 0, 0, 0]
- Name: Out
Format: Float32
Stride: 16
FillSize: 48
- Name: ExpectedOut # The result we expect
Format: Float32
Stride: 16
Data: [ nan, nan, 0, 0, 0, .5, nan, 0, 0.718280, 0.599976, 0.611, nan,]
# NaN, NaN, 0, 0, 0, .5, NaN, 0, 0.718280, 0.599976, 0.611,
Data: [ 0, 0, 0, 0.5, 0, 0.71828, 0.599976, 0.611, 0, 0, 0, 0 ]
Results:
- Result: Test1
Rule: BufferFloatEpsilon
Expand Down
16 changes: 9 additions & 7 deletions test/Feature/HLSLLib/log10.16.test
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,18 +24,20 @@ Shaders:
Buffers:
- Name: In
Format: Float16
Stride: 8
Data: [ 0x7e00, 0xfc00, 0x8001, 0x8000, 0x0000, 0x0001, 0x7c00, 0x3c00, 0xbc00, 0x4900, 0x2e66, 0x5640,]
# NaN, -Inf, -denorm, -0, 0, denorm, Inf, 1, -1, 10, 0.1, 100,
Stride: 0
Data: [0x8000, 0x0000, 0x3c00, 0xbc00, 0x4900, 0x2e66, 0x5640,]
# [-0.0, 0.0, 1.0, -1.0, 10.0, 0.0960693359375, 100.0]
- Name: Out
Format: Float16
Stride: 8
Stride: 0
FillSize: 24
- Name: ExpectedOut # The result we expect
Format: Float16
Stride: 8
Data: [ 0x7e00, 0x7e00, 0x7e00, 0xfc00, 0xfc00, 0xc73a, 0x7c00, 0x0000, 0x7e00, 0x3c00, 0xbc00, 0x4000,]
# NaN, NaN, NaN, -inf, -Inf, -7.2265, inf, 0, nan, 1, -1, 2,
Stride: 0
Data: [ 0xFC00, 0xFC00, 0x0000, 0x7FFF, 0x3BFF, 0xBC00, 0x3FFF, 0xFC00, 0xFC00, 0xFC00, 0xFC00, 0xFC00 ]
# [-inf, -inf, 0.0, NaN, 0.999755859375, -1.0, 1.999755859375, -inf, -inf, -inf, -inf, -inf]


Results:
- Result: Test1
Rule: BufferFloatULP
Expand Down
11 changes: 4 additions & 7 deletions test/Feature/HLSLLib/mad.32.test
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -45,18 +45,15 @@ Buffers:
- Name: M0
Format: Float32
Stride: 16
Data: [ NaN, -Inf, -0x1.e7d42cp-127, -0, 0, 0x1.e7d42cp-127, Inf, 1.0, -1.0, 0, 1, 1.5 ]
# NaN, -Inf, -denorm, -0, 0, denorm, Inf, 1.0, -1.0, 0, 1, 1.5
Data: [ -0, 0, 1.0, -1.0, 0, 1, 1.5 ]
- Name: A0
Format: Float32
Stride: 16
Data: [ NaN, -Inf, -0x1.e7d42cp-127, -0, 0, 0x1.e7d42cp-127, Inf, 1.0, -1.0, 0, 1, 10 ]
# NaN, -Inf, -denorm, -0, 0, denorm, Inf, 1.0, -1.0, 0, 1, 10
Data: [ -0, 0, 1.0, -1.0, 0, 1, 10 ]
- Name: B0
Format: Float32
Stride: 16
Data: [ NaN, -Inf, -0x1.e7d42cp-127, -0, 0, 0x1.e7d42cp-127, Inf, 1.0, -1.0, 1, 0, -5.5 ]
# NaN, -Inf, -denorm, -0, 0, denorm, Inf, 1.0, -1.0, 1, 0, -5.5
Data: [ -0, 0, 1.0, -1.0, 1, 0, -5.5 ]
- Name: M1
Format: Int32
Stride: 16
Expand Down Expand Up @@ -88,7 +85,7 @@ Buffers:
- Name: ExpectedOut0
Format: Float32
Stride: 16
Data: [ NaN, NaN, 0, 0, 0, 0, Inf, 2, 0, 1, 1, 9.5, 2, 9.5, Inf, -Inf ]
Data: [ 0, 0, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 2, 9.5, inf, -inf ]
- Name: Out1
Format: Int32
Stride: 16
Expand Down
8 changes: 4 additions & 4 deletions test/Feature/HLSLLib/sinh.16.test
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,17 +25,17 @@ Buffers:
- Name: In
Format: Float16
Stride: 8
Data: [ 0x7e00, 0xfc00, 0x8001, 0x8000, 0x0000, 0x0001, 0x7c00, 0x3c00, 0xbc00, 0x7e00, 0x7e00, 0x7e00,]
# NaN, -Inf, -denorm, -0, 0, denorm, Inf, 1, -1,
Data: [0x8000, 0x0000, 0x3c00, 0xbc00, 0x8000, 0x0000, 0x3c00, 0xbc00, 0x8000, 0x0000, 0x3c00, 0xbc00]
# -0, 0, 1, -1, -0, 0, 1, -1, -0, 0, 1, -1,
- Name: Out
Format: Float16
Stride: 8
FillSize: 24
- Name: ExpectedOut # The result we expect
Format: Float16
Stride: 8
Data: [ 0x7e00, 0xfc00, 0x0000, 0x0000, 0x0000, 0x0000, 0x7c00, 0x3cb3, 0xbcb3, 0x7e00, 0x7e00, 0x7e00,]
# NaN, -Inf, 0.0, 0.0, 0.0, 0.0, Inf, 1.175201, -1.175201,
Data: [ 0x0, 0x0, 0x3CB3, 0xBCB3, 0x0, 0x0, 0x3CB3, 0xBCB3, 0x0, 0x0, 0x3CB3, 0xBCB3 ]
# -0, 0, 1.175, -1.175, -0, 0, 1.175, -1.175, -0, 0, 1.175, -1.175
Results:
- Result: Test1
Rule: BufferFloatULP
Expand Down
Loading