Timeline


and .

08/19/10:

03:06 Ticket #27 (fold, foldSeg and scan fail to compile for tuple types) created by seanl
The fold.inl fails to compile when TyOut is a struct. This is …

08/18/10:

06:39 Ticket #26 (Internal error in filter test with criterion) created by chak
On my MBP with a NVIDIA GeForce 9400M (256MB VRAM) and CUDA 3.1, I …
06:27 Ticket #15 (Support for stencil-based traversal in the language front-end) closed by chak
fixed: I implemented the following:[…]
04:40 Ticket #19 (CUDA backend doesn't support shapes) closed by tmcdonell
fixed
04:40 Ticket #6 (CUDA backend doesn't support slice) closed by tmcdonell
fixed: […]
04:40 Ticket #5 (CUDA backend doesn't support replicate) closed by tmcdonell
fixed: […]

08/16/10:

02:22 Ticket #20 (Code generation for nested tuple projection fails) closed by tmcdonell
fixed: […]

08/12/10:

00:11 Ticket #17 (Add atan2 to the front end) closed by tmcdonell
fixed: […]

08/11/10:

15:51 Ticket #25 (fold produces incorrect result when a non-identity value is given as the ...) closed by tmcdonell
duplicate: Merge with #23

08/10/10:

10:43 Ticket #21 (Texture binding for IndexScalar does not work properly with the projection ...) closed by tmcdonell
fixed: […]
10:42 Ticket #22 (IndexScalar does not produce the correct value when it is used in the ...) closed by tmcdonell
fixed: […]
09:57 Ticket #24 (fclabels > 0.9.1 incompatibility issue.) closed by tmcdonell
fixed: Fixed.[…]
03:36 Ticket #25 (fold produces incorrect result when a non-identity value is given as the ...) created by seanl
For example, […]

08/09/10:

14:36 Ticket #24 (fclabels > 0.9.1 incompatibility issue.) created by seanl
The fclabels package has been updated. mkLabels has been renamed to …
14:12 Ticket #23 (Support non-identity elements for scan, fold) created by seanl
It seems that the default value is ignored for {{{Prelude.snd $ …
08:27 Ticket #22 (IndexScalar does not produce the correct value when it is used in the ...) created by seanl
IndexScalar does not produce the correct value when it is used in …
07:23 Ticket #21 (Texture binding for IndexScalar does not work properly with the projection ...) created by seanl
The execution of the code fragment […]

08/06/10:

05:36 Ticket #20 (Code generation for nested tuple projection fails) created by seanl
The CUDA backend appears to be failing when generating the code for the …

08/02/10:

08:40 Ticket #7 (CUDA backend doesn't support foldSeg) closed by tmcdonell
fixed

07/28/10:

06:39 Ticket #19 (CUDA backend doesn't support shapes) created by tmcdonell

07/24/10:

11:16 Ticket #18 (Permute does not properly write-combine results) created by tmcdonell
When one or more threads try to write to the same location, the hardware …
04:01 Ticket #16 (CUDA: correct C version of function not emitted for Double types) closed by tmcdonell
invalid: The 'ccall' helper function is for calling primitive C functions such as …

07/23/10:

09:20 Ticket #17 (Add atan2 to the front end) created by rmukhtar
Currently atan2 is missing from the front end language.
08:58 Ticket #16 (CUDA: correct C version of function not emitted for Double types) created by rmukhtar
In the file "Data/Array/Accelerate/CUDA/CodeGen.hs" the patterns for …
01:52 Ticket #4 (scanl is buggy in the CUDA backend) closed by tmcdonell
fixed

07/22/10:

06:13 Ticket #13 (Internal state structures should use HashTable) closed by tmcdonell
fixed
06:13 Ticket #14 (Module caching is lost in the presence of free array variables) closed by tmcdonell
fixed
03:44 Ticket #15 (Support for stencil-based traversal in the language front-end) created by blever
A "stencil" primitive would be a very useful in the language front-end of …
00:21 Ticket #14 (Module caching is lost in the presence of free array variables) created by tmcdonell
The hash tags generated for array computations in D.A.A.CUDA.Analysis.Hash …
00:14 Ticket #13 (Internal state structures should use HashTable) created by tmcdonell
The internal state structures in D.A.A.CUDA.State use Data.Map to track …
00:10 Ticket #12 (Unused array computations lead to memory leaks) created by tmcdonell
The CUDA backend uses explicit reference tracking for device arrays. …

07/20/10:

04:53 Ticket #11 (Hash tag for CUDA version of backpermute does not include projection ...) closed by tmcdonell
fixed
04:52 Ticket #9 (The code generation fails when dealing with the operations on arrays of ...) closed by tmcdonell
fixed
01:38 Ticket #11 (Hash tag for CUDA version of backpermute does not include projection ...) created by rmukhtar
accToKey implementation for Backpermute (line 38 of …
Note: See TracTimeline for information about the timeline view.