Skip to content

Commit 7a57ae0

Browse files
author
bragadeesh
committed
updating for release of v2.12
1 parent b7da307 commit 7a57ae0

File tree

3 files changed

+27
-13
lines changed

3 files changed

+27
-13
lines changed

README.md

+6-10
Original file line numberDiff line numberDiff line change
@@ -15,16 +15,11 @@ Pre-built binaries are available [here][binary_release].
1515

1616
## What's New
1717

18-
- Support for power-of-7 size transforms
19-
- Pre-callback & post-callback feature that enables custom pre-processing
20-
of input/output data directly by the library with user callback function
18+
- Support for powers of 11&13 size transforms
2119
- Support for 1D large size transforms with no extra memory allocation
22-
requirement for certain sizes
23-
- Significant uplift of 1D complex transform performance
24-
- Significant uplift of 1D real transform performance for power-of-2 sizes
25-
- 1D large size limit relaxation for complex transforms
26-
- 2D/3D size limit relaxation on real and complex transforms
27-
- Binary caching feature
20+
requirement with environment flag CLFFT_REQUEST_LIB_NOMEMALLOC=1
21+
for complex FFTs of powers of 2,3,5,10 sizes
22+
2823

2924
## Note
3025

@@ -59,7 +54,8 @@ discrete Fast Fourier Transforms. The library:
5954
interleaved (real and complex components as a pair contiguous in
6055
memory) formats.
6156

62-
- supports dimension lengths that can be any combination of powers of 2, 3, 5, and 7.
57+
- supports dimension lengths that can be any combination of powers of
58+
2, 3, 5, 7, 11 and 13.
6359

6460
- Supports single and double precision floating point formats.
6561

ReleaseNotes.txt

+19-1
Original file line numberDiff line numberDiff line change
@@ -13,12 +13,30 @@ implementation of discrete Fast Fourier Transforms. It:
1313
separate arrays) and interleaved (real and complex
1414
components as a pair contiguous in memory) formats
1515
* Supports dimension lengths that can be any mix of
16-
powers of 2, 3, 5 and 7
16+
powers of 2, 3, 5, 7, 11 and 13
1717
* Supports single and double precision floating-point
1818
formats
1919
* Supports in-place or out-of-place transforms
2020

2121

22+
clFFT - Release Notes - version 2.12.0
23+
--------------------------------------
24+
25+
New features of this release:
26+
27+
* Addition of radix 11 & 13 enables support for powers of 11 & 13 size
28+
transforms
29+
* Support for 1D large size transforms with no extra memory allocation
30+
requirement with environment flag CLFFT_REQUEST_LIB_NOMEMALLOC=1
31+
for complex FFTs of powers of 3,5,10 sizes. Note that support for
32+
power-of-2 size has been available since version 2.10.
33+
34+
Driver notes:
35+
36+
* This library version has been tested with Catalyst
37+
Pro driver version 15.201 on Firepro W9100.
38+
39+
2240
clFFT - Release Notes - version 2.10.2
2341
--------------------------------------
2442

src/CMakeLists.txt

+2-2
Original file line numberDiff line numberDiff line change
@@ -40,11 +40,11 @@ if( NOT DEFINED CLFFT_VERSION_MAJOR )
4040
endif( )
4141

4242
if( NOT DEFINED CLFFT_VERSION_MINOR )
43-
set( CLFFT_VERSION_MINOR 10 )
43+
set( CLFFT_VERSION_MINOR 12 )
4444
endif( )
4545

4646
if( NOT DEFINED CLFFT_VERSION_PATCH )
47-
set( CLFFT_VERSION_PATCH 2 )
47+
set( CLFFT_VERSION_PATCH 0 )
4848
endif( )
4949

5050
set( CLFFT_VERSION "${CLFFT_VERSION_MAJOR}.${CLFFT_VERSION_MINOR}.${CLFFT_VERSION_PATCH}")

0 commit comments

Comments
 (0)