On Sunday, 18 August 2013 at 05:26:00 UTC, Manu wrote:
movups is not good. It'll be a lot faster (and portable) if you
use movaps.
Process looks something like:
* do the first few from a[0] until a's alignment interval as
scalar
* load the left of b's aligned pair
* loop for each aligne
movups is not good. It'll be a lot faster (and portable) if you use movaps.
Process looks something like:
* do the first few from a[0] until a's alignment interval as scalar
* load the left of b's aligned pair
* loop for each aligned vector in a
- load a[n..n+4] aligned
- load the ri
On Sunday, 18 August 2013 at 05:07:12 UTC, Manu wrote:
On 18 August 2013 14:39, Ilya Yaroshenko
wrote:
On Sunday, 18 August 2013 at 01:53:53 UTC, Manu wrote:
It doesn't look like you account for alignment.
This is basically not-portable (I doubt unaligned loads in
this context
are
faster
On 18 August 2013 14:39, Ilya Yaroshenko wrote:
> On Sunday, 18 August 2013 at 01:53:53 UTC, Manu wrote:
>
>> It doesn't look like you account for alignment.
>> This is basically not-portable (I doubt unaligned loads in this context
>> are
>> faster than performing scalar operations), and possibl
On Saturday, 17 August 2013 at 19:38:52 UTC, John Colvin wrote:
On Saturday, 17 August 2013 at 19:24:52 UTC, Ilya Yaroshenko
wrote:
BTW: -march=native automatically implies -mtune=native
Thanks, I`ll remove mtune)
It would be really interesting if you could try writing the
same code in c, b
On Sunday, 18 August 2013 at 01:53:53 UTC, Manu wrote:
It doesn't look like you account for alignment.
This is basically not-portable (I doubt unaligned loads in this
context are
faster than performing scalar operations), and possibly
inefficient on x86
too.
dotProduct uses unaligned loads (
On Saturday, 16 March 2013 at 14:42:58 UTC, Suliman wrote:
Hi folks! I had wrote small article about Rust vs D. I hope
that you will like it!
http://versusit.org/rust-vs-d
I don't know if you still follow this, but there's a typo here:
Now, let’s see, how the code, outputting the word "Hello
It doesn't look like you account for alignment.
This is basically not-portable (I doubt unaligned loads in this context are
faster than performing scalar operations), and possibly inefficient on x86
too.
To make it account for potentially random alignment will be awkward, but it
might be possible t
On Saturday, 17 August 2013 at 20:12:49 UTC, Paul Jurczak wrote:
On Saturday, 17 August 2013 at 11:00:20 UTC, John Colvin wrote:
On Saturday, 17 August 2013 at 09:21:27 UTC, Paul Jurczak
wrote:
On Saturday, 17 August 2013 at 01:05:18 UTC, Tyler Jameson
Little wrote:
[..]
It's true though that
Ilya Yaroshenko:
http://spiceandmath.blogspot.ru/2013/08/simd-implementation-of-dot-product_17.html
From the blog post:
Compile fast_math code from other program separately and then
link it. This is easy solution. However this is a step back to
C.<
To introduce a @fast_math attribute. This i
On Saturday, 17 August 2013 at 11:00:20 UTC, John Colvin wrote:
On Saturday, 17 August 2013 at 09:21:27 UTC, Paul Jurczak wrote:
On Saturday, 17 August 2013 at 01:05:18 UTC, Tyler Jameson
Little wrote:
[..]
It's true though that there are much better answers (and
questions) here than on SO, an
On Saturday, 17 August 2013 at 19:24:52 UTC, Ilya Yaroshenko
wrote:
BTW: -march=native automatically implies -mtune=native
Thanks, I`ll remove mtune)
It would be really interesting if you could try writing the same
code in c, both a scalar version and a version using gcc's vector
instrinsic
BTW: -march=native automatically implies -mtune=native
Thanks, I`ll remove mtune)
On Saturday, 17 August 2013 at 18:50:15 UTC, Ilya Yaroshenko
wrote:
http://spiceandmath.blogspot.ru/2013/08/simd-implementation-of-dot-product_17.html
Ilya
Nice, that's a good speedup.
BTW: -march=native automatically implies -mtune=native
http://spiceandmath.blogspot.ru/2013/08/simd-implementation-of-dot-product_17.html
Ilya
On Saturday, 17 August 2013 at 15:22:26 UTC, Carl Sturtivant
wrote:
Hello Andrei, Ali,
Well, CS2 in D is over, and grades are in. Being a summer
course it was conducted at twice the speed as in a regular
semester.
[...]
That is truly awesome news !
What happened is that I unclear what the state of play is; I
am not
asserting the wrongness of TDPL. Still, I am reassured by your
responses.
Hi Carl -- TDPL is for the most part in good shape. There are a
few inaccuracies, but I'd say most concern corner cases that
are unlike to deter the le
On 16/08/2013 20:14, Bruno Medeiros wrote:
Found a bug in the meanwhile, the DDoc viewer is not rendering $(D) DDoc
macros.
Speaking of which, where does that macro come from? Seems similar to the
D_CODE macro, but it's not documented as one of the default DDoc macros,
nor could I find it as
On 16/08/2013 20:22, Andrei Alexandrescu wrote:
On 8/16/13 12:14 PM, Bruno Medeiros wrote:
I've updated the Features wiki with new screenshots, and revised the
text to be more clear:
http://code.google.com/p/ddt/wiki/Features
(like removing the "A JDT-like project model" references which actuall
On 16/08/2013 22:05, Piotr Szturmaj wrote:
I've pointed path to the DMD compiler executable but the library paths
weren't filled automatically. Maybe it's a bug? (eclipse CDT Kepler)
Yeah, it is a bug. It was a new feature in this release, but I made a
mistake creating the test case (and then
On 16/08/2013 12:08, Jacob Carlborg wrote:
On 2013-08-15 21:20, Bruno Medeiros wrote:
A new version of DDT - D Development tools is out.
The major change is the new parser which is updated to the latest
version of D, and is much more robust than the previous one.
Full changelog/info here:
https
On Saturday, 17 August 2013 at 09:21:27 UTC, Paul Jurczak wrote:
On Saturday, 17 August 2013 at 01:05:18 UTC, Tyler Jameson
Little wrote:
[..]
It's true though that there are much better answers (and
questions) here than on SO, and I'm beginning to shift my
search from Google to the forum sear
On Saturday, 17 August 2013 at 01:05:18 UTC, Tyler Jameson Little
wrote:
[..]
It's true though that there are much better answers (and
questions) here than on SO, and I'm beginning to shift my
search from Google to the forum search, but this isn't
something a newcomer will know to do, especial
23 matches
Mail list logo