Re: drastic slowdown for copies
On Friday, 29 May 2015 at 07:51:31 UTC, thedeemon wrote: On Thursday, 28 May 2015 at 21:23:11 UTC, Momo wrote: Ah, actually it's more complicated, as it depends on inlining a lot. Yes. And real functions are more complex and inlining is no reliable option. Indeed, without -O and -inline I was able to get by_ref to be slightly slower than by_copy for struct of 4 ints. But when inlining turns on, the numbers change in different directions. And for 5 ints inlining influence is quite different: 4 ints: 5 ints: -release by ref: 53 by ref: 53 by copy: 57 by copy: 137 by move: 54 by move: 137 -release -O by ref: 38 by ref: 34 by copy: 54 by copy: 137 by move: 49 by move: 137 -release -O -inline by ref: 15 by ref: 20 by copy: 72 by copy: 91 by move: 72 by move: 91 So as you can see, it is 2-3 times slower. Is there an alternative?
Re: Windows Universal/Store apps support
On Friday, 29 May 2015 at 03:23:39 UTC, Rikki Cattermole wrote: On 29/05/2015 3:57 a.m., Olivier Prince wrote: I searched the forum to find if there is some support for new Windows development technologies and I didn't find anything related (except some rants about WinRT 3 years ago). - Is there any support in D or phobos for developping this kind of applications? No. - Does D support ARM as Windows target? Yes/No. ldc/gdc guys probably are a good place to start. - Are there any plans to support the specific libraries in this respect (something like C++/CX extensions or COM metadata files)? What exactly do you need? WinRT is an evolution of COM with a .NET feel to the API. Basically it is COM where objects also need to implement a new interface, IInspectable, and .NET metadata stored in .winmd files is used instead of COM type libraries. In a way WinRT is the return of Ext-VOS, the original idea behind .NET. Most likely caused by the Longhorn failure to write everything in .NET, as most new Windows APIs have been introduced as COM components since Vista. So any language targeting WinRT, or Universal Apps as they are now known, needs to to support COM alongside the required interfaces for interoperability between languages and be able to consume/produce .NET metadata files. -- Paulo
Re: drastic slowdown for copies
On Friday, 29 May 2015 at 07:51:31 UTC, thedeemon wrote: Above was on Core 2 Quad, here's for Core i3: 4 ints 5 ints -release by ref: 67 by ref: 66 by copy: 44 by copy: 142 by move: 45 by move: 137 -release -O by ref: 29 by ref: 29 by copy: 41 by copy: 141 by move: 40 by move: 142 -release -O -inline by ref: 16 by ref: 20 by copy: 83 by copy: 104 by move: 83 by move: 104
Re: Windows Universal/Store apps support
On 29/05/2015 7:03 p.m., Paulo Pinto wrote: On Friday, 29 May 2015 at 03:23:39 UTC, Rikki Cattermole wrote: On 29/05/2015 3:57 a.m., Olivier Prince wrote: I searched the forum to find if there is some support for new Windows development technologies and I didn't find anything related (except some rants about WinRT 3 years ago). - Is there any support in D or phobos for developping this kind of applications? No. - Does D support ARM as Windows target? Yes/No. ldc/gdc guys probably are a good place to start. - Are there any plans to support the specific libraries in this respect (something like C++/CX extensions or COM metadata files)? What exactly do you need? WinRT is an evolution of COM with a .NET feel to the API. Basically it is COM where objects also need to implement a new interface, IInspectable, and .NET metadata stored in .winmd files is used instead of COM type libraries. In a way WinRT is the return of Ext-VOS, the original idea behind .NET. Most likely caused by the Longhorn failure to write everything in .NET, as most new Windows APIs have been introduced as COM components since Vista. So any language targeting WinRT, or Universal Apps as they are now known, needs to to support COM alongside the required interfaces for interoperability between languages and be able to consume/produce .NET metadata files. -- Paulo D already supports COM. For .winmd, you should be good to go for -m32mscoff and -m64! Just add LFLAGS=/winmd Although this really really needs to be tested.
Re: drastic slowdown for copies
On Thursday, 28 May 2015 at 21:23:11 UTC, Momo wrote: Ah, actually it's more complicated, as it depends on inlining a lot. Indeed, without -O and -inline I was able to get by_ref to be slightly slower than by_copy for struct of 4 ints. But when inlining turns on, the numbers change in different directions. And for 5 ints inlining influence is quite different: 4 ints: 5 ints: -release by ref: 53 by ref: 53 by copy: 57 by copy: 137 by move: 54 by move: 137 -release -O by ref: 38 by ref: 34 by copy: 54 by copy: 137 by move: 49 by move: 137 -release -O -inline by ref: 15 by ref: 20 by copy: 72 by copy: 91 by move: 72 by move: 91
Re: drastic slowdown for copies
On Thursday, 28 May 2015 at 21:23:11 UTC, Momo wrote: I'm currently investigating the difference of speed between references and copies. And it seems that copies got a immense slowdown if they reach a size of = 20 bytes. This is processor-specific, on different models of CPUs you might get different results. Here's what I see running your program with 4 and 5 ints in the struct: C:\prog\Ddmd copyref.d -ofcopyref.exe -release -O -inline 16u C:\prog\Dcopyref.exe by ref: 18 by copy: 85 by move: 84 C:\prog\Dcopyref.exe by ref: 18 by copy: 72 by move: 72 C:\prog\Dcopyref.exe by ref: 16 by copy: 72 by move: 72 C:\prog\Ddmd copyref.d -ofcopyref.exe -release -O -inline 20u C:\prog\Dcopyref.exe by ref: 23 by copy: 98 by move: 91 C:\prog\Dcopyref.exe by ref: 20 by copy: 91 by move: 102 C:\prog\Dcopyref.exe by ref: 23 by copy: 91 by move: 91 I see these digits on an old Core 2 Quad and very similar on a Core i3. So your findings are not reproducible.
Re: Windows Universal/Store apps support
On Friday, 29 May 2015 at 10:01:53 UTC, Rikki Cattermole wrote: On 29/05/2015 9:55 p.m., Paulo Pinto wrote: On Friday, 29 May 2015 at 07:41:14 UTC, Rikki Cattermole wrote: On 29/05/2015 7:03 p.m., Paulo Pinto wrote: On Friday, 29 May 2015 at 03:23:39 UTC, Rikki Cattermole wrote: On 29/05/2015 3:57 a.m., Olivier Prince wrote: I searched the forum to find if there is some support for new Windows development technologies and I didn't find anything related (except some rants about WinRT 3 years ago). - Is there any support in D or phobos for developping this kind of applications? No. - Does D support ARM as Windows target? Yes/No. ldc/gdc guys probably are a good place to start. - Are there any plans to support the specific libraries in this respect (something like C++/CX extensions or COM metadata files)? What exactly do you need? WinRT is an evolution of COM with a .NET feel to the API. Basically it is COM where objects also need to implement a new interface, IInspectable, and .NET metadata stored in .winmd files is used instead of COM type libraries. In a way WinRT is the return of Ext-VOS, the original idea behind .NET. Most likely caused by the Longhorn failure to write everything in .NET, as most new Windows APIs have been introduced as COM components since Vista. So any language targeting WinRT, or Universal Apps as they are now known, needs to to support COM alongside the required interfaces for interoperability between languages and be able to consume/produce .NET metadata files. -- Paulo D already supports COM. For .winmd, you should be good to go for -m32mscoff and -m64! Just add LFLAGS=/winmd Although this really really needs to be tested. As far as I am aware D supports old style COM, not COM with WinRT improvements. I don't get your hint with LFLAGS. https://msdn.microsoft.com/en-us/library/jj157268.aspx So unless this isn't doing what I think it is doing, then yes it is relevant. No guarantees about what it requires for input info regarding e.g. symbols however. Are you aware what are .NET metadata files? I can't say anything about that. Again try the flag. If it doesn't do what I'm expecting it does, no harm. It's just a feature that definitely isn't implemented or supported by the toolchain. Of course, just because it generates the file, doesn't mean that there will be anything of use within it. The compiler provides the required information in C++/CX mode. When using pure C++ with the Windows Runtime C++ Template Library instead[0], it requires an idl file to be given to the MIDL compiler, that is then fed to the linker, like old style COM type libraries. [0] - https://msdn.microsoft.com/en-us/library/windows/apps/hh438466.aspx [1] - https://msdn.microsoft.com/en-us/library/windows/apps/hh973463.aspx
Mutable reference for immutable AA
I need mutable storage for immutable associative array. Just create new immutable AA and store it for future passing it between threads/fibers. First attempt: just immutable AA immutable aa = [1:1, 2:1]; aa = [1:1, 2:1]; // fail, can't assign a new AA Second attempt: mutable AA with immutable elements. immutable (int)[string] aa = [1:1, 2:1]; aa = [1:1, 2:1]; // sucess! aa[2] = 2; // doesn't compile, is AA immutable? aa.remove(1); // fail, this compiles, AA actually isn't immutable Third attempt: Rebindable Rebindable!(immutable int[string]) aa; // fail, Rebindable doesn't accept AAs
Re: Mutable reference for immutable AA
I made trivial pull request - https://github.com/D-Programming-Language/phobos/pull/3341 RebindableAA!(immutable int[string]) aa = [a: 1, b: 2]; // works assert(aa[a] == 1); // cool aa = [a: 3, b: 4]; // nice auto bb = aa; // yes bb = [a: 4, b: 5]; // super aa[a] = 2; // error, good aa.remove(a);// error, very good
Re: Windows Universal/Store apps support
On 29/05/2015 9:55 p.m., Paulo Pinto wrote: On Friday, 29 May 2015 at 07:41:14 UTC, Rikki Cattermole wrote: On 29/05/2015 7:03 p.m., Paulo Pinto wrote: On Friday, 29 May 2015 at 03:23:39 UTC, Rikki Cattermole wrote: On 29/05/2015 3:57 a.m., Olivier Prince wrote: I searched the forum to find if there is some support for new Windows development technologies and I didn't find anything related (except some rants about WinRT 3 years ago). - Is there any support in D or phobos for developping this kind of applications? No. - Does D support ARM as Windows target? Yes/No. ldc/gdc guys probably are a good place to start. - Are there any plans to support the specific libraries in this respect (something like C++/CX extensions or COM metadata files)? What exactly do you need? WinRT is an evolution of COM with a .NET feel to the API. Basically it is COM where objects also need to implement a new interface, IInspectable, and .NET metadata stored in .winmd files is used instead of COM type libraries. In a way WinRT is the return of Ext-VOS, the original idea behind .NET. Most likely caused by the Longhorn failure to write everything in .NET, as most new Windows APIs have been introduced as COM components since Vista. So any language targeting WinRT, or Universal Apps as they are now known, needs to to support COM alongside the required interfaces for interoperability between languages and be able to consume/produce .NET metadata files. -- Paulo D already supports COM. For .winmd, you should be good to go for -m32mscoff and -m64! Just add LFLAGS=/winmd Although this really really needs to be tested. As far as I am aware D supports old style COM, not COM with WinRT improvements. I don't get your hint with LFLAGS. https://msdn.microsoft.com/en-us/library/jj157268.aspx So unless this isn't doing what I think it is doing, then yes it is relevant. No guarantees about what it requires for input info regarding e.g. symbols however. Are you aware what are .NET metadata files? I can't say anything about that. Again try the flag. If it doesn't do what I'm expecting it does, no harm. It's just a feature that definitely isn't implemented or supported by the toolchain. Of course, just because it generates the file, doesn't mean that there will be anything of use within it.
Re: Windows Universal/Store apps support
On Friday, 29 May 2015 at 03:23:39 UTC, Rikki Cattermole wrote: On 29/05/2015 3:57 a.m., Olivier Prince wrote: I searched the forum to find if there is some support for new Windows development technologies and I didn't find anything related (except some rants about WinRT 3 years ago). - Is there any support in D or phobos for developping this kind of applications? No. - Does D support ARM as Windows target? Yes/No. ldc/gdc guys probably are a good place to start. - Are there any plans to support the specific libraries in this respect (something like C++/CX extensions or COM metadata files)? What exactly do you need? Language integrated reference counting, runtime templates, partial COM classes, strings, arrays, collections, exceptions, events, delegates, attributes. I know that D has these features but I doubt that thay are mapped to COM counterparts. - At least, is there a skeleton Store app written in D? No.
Re: Windows Universal/Store apps support
On Friday, 29 May 2015 at 07:41:14 UTC, Rikki Cattermole wrote: On 29/05/2015 7:03 p.m., Paulo Pinto wrote: On Friday, 29 May 2015 at 03:23:39 UTC, Rikki Cattermole wrote: On 29/05/2015 3:57 a.m., Olivier Prince wrote: I searched the forum to find if there is some support for new Windows development technologies and I didn't find anything related (except some rants about WinRT 3 years ago). - Is there any support in D or phobos for developping this kind of applications? No. - Does D support ARM as Windows target? Yes/No. ldc/gdc guys probably are a good place to start. - Are there any plans to support the specific libraries in this respect (something like C++/CX extensions or COM metadata files)? What exactly do you need? WinRT is an evolution of COM with a .NET feel to the API. Basically it is COM where objects also need to implement a new interface, IInspectable, and .NET metadata stored in .winmd files is used instead of COM type libraries. In a way WinRT is the return of Ext-VOS, the original idea behind .NET. Most likely caused by the Longhorn failure to write everything in .NET, as most new Windows APIs have been introduced as COM components since Vista. So any language targeting WinRT, or Universal Apps as they are now known, needs to to support COM alongside the required interfaces for interoperability between languages and be able to consume/produce .NET metadata files. -- Paulo D already supports COM. For .winmd, you should be good to go for -m32mscoff and -m64! Just add LFLAGS=/winmd Although this really really needs to be tested. As far as I am aware D supports old style COM, not COM with WinRT improvements. I don't get your hint with LFLAGS. Are you aware what are .NET metadata files? .. Paulo
Re: drastic slowdown for copies
Perhaps you can give me another detailed answer.
I get a slowdown for all parts (ref, copy and move) if I use
uninitialized floats. I got these results from the following code:
by ref: 2369
by copy: 2335
by move: 2341
Code:
struct vec2f {
float x;
float y;
}
But if I assign 0 to them I got these results:
by ref: 49
by copy: 22
by move: 25
Why?
Re: Get index of string in array at compile time
On 05/29/2015 06:43 PM, tcak wrote:
I have define an immutable string array:
[code]
immutable string[] placeHolderDefinitionList = [
!-- fullname --,
!-- list item --
];
[/code]
I need to get index of a string at compile time. So I have written a
function as below:
[code]
public size_t getPlaceholderIndex(string PlaceHolderText)( size_t
lookIndex = 0 ){
static if( lookIndex = placeHolderDefinitionList.length )
return size_t.max;
static if( placeHolderDefinitionList[lookIndex] == PlaceHolderText )
return lookIndex;
else
return getPlaceholderIndex( lookIndex + 1 );
}
[/code]
Called it as below:
[code]
writeln( Index: , getPlaceholderIndex!!-- list item --( 0 ) );
[/code]
Compiler didn't like my code. It said those things to my face:
main.d(22): Error: variable lookIndex cannot be read at compile time
main.d(25): Error: variable lookIndex cannot be read at compile time
main.d(28): Error: template main.getPlaceholderIndex cannot deduce
function from argument types !()(ulong), candidates are:
main.d(21):main.getPlaceholderIndex(string
PlaceHolderText)(size_t lookIndex = 0)
main.d(105): Error: template instance main.getPlaceholderIndex!!--
list item -- error instantiating
This is my first try with this type of code. So I am confused a little
about what I should be doing.
The easiest way is to just write runtime code and evaluate it via CTFE:
immutable string[] placeHolderDefinitionList = [
!-- fullname --,
!-- list item --
];
void main(){
import std.stdio, std.algorithm;
enum index=placeHolderDefinitionList.countUntil(!-- list item --);
writeln(Index: , index);
}
If you never need the placeHolderDefinitionList at runtime, make it enum
instead of immutable.
Get index of string in array at compile time
I have define an immutable string array:
[code]
immutable string[] placeHolderDefinitionList = [
!-- fullname --,
!-- list item --
];
[/code]
I need to get index of a string at compile time. So I have
written a function as below:
[code]
public size_t getPlaceholderIndex(string PlaceHolderText)( size_t
lookIndex = 0 ){
static if( lookIndex = placeHolderDefinitionList.length )
return size_t.max;
static if( placeHolderDefinitionList[lookIndex] ==
PlaceHolderText )
return lookIndex;
else
return getPlaceholderIndex( lookIndex + 1 );
}
[/code]
Called it as below:
[code]
writeln( Index: , getPlaceholderIndex!!-- list item --( 0 )
);
[/code]
Compiler didn't like my code. It said those things to my face:
main.d(22): Error: variable lookIndex cannot be read at compile
time
main.d(25): Error: variable lookIndex cannot be read at compile
time
main.d(28): Error: template main.getPlaceholderIndex cannot
deduce function from argument types !()(ulong), candidates are:
main.d(21):main.getPlaceholderIndex(string
PlaceHolderText)(size_t lookIndex = 0)
main.d(105): Error: template instance
main.getPlaceholderIndex!!-- list item -- error instantiating
This is my first try with this type of code. So I am confused a
little about what I should be doing.
Re: Windows Universal/Store apps support
On Thursday, 28 May 2015 at 15:57:42 UTC, Olivier Prince wrote: I searched the forum to find if there is some support for new Windows development technologies and I didn't find anything related (except some rants about WinRT 3 years ago). - Is there any support in D or phobos for developping this kind of applications? - Does D support ARM as Windows target? - Are there any plans to support the specific libraries in this respect (something like C++/CX extensions or COM metadata files)? - At least, is there a skeleton Store app written in D? Thanks. I don't know of any work outside of the old Microsoft slides, so likely someone who is familiar with writing a C++ app will need to try it out and request improvements where things are missing. Juno may be a good place to go for COM. https://github.com/JesseKPhillips/Juno-Windows-Class-Library
Re: Windows Universal/Store apps support
On Thursday, 28 May 2015 at 15:57:42 UTC, Olivier Prince wrote: [snip] There isn't yet a polished alternative to MS VS Windows Store toolchain, but probably you don't need most of it (e.g. you can have a C++/XAML app that calls you D code). I noticed that Vibe.d has some WinRT support. Here's their WinRT driver, which uses calls some parts of WinRT API: https://github.com/rejectedsoftware/vibe.d/blob/master/source/vibe/core/drivers/winrt.d
Mixin - to get to the content-type `MapResult!(__lambda1, int[]).MapResult`
Hi,
This code prints the arrays:
[5]
[6]
[7]
import std.stdio, std.algorithm;
static int idx;
void walk(R)(R range) {
while (!range.empty) {
range.front;
range.popFront;
++idx;
}
}
void main() {
[5, 6, 7].map!(a = [a].writeln).walk;
}
How should I apply mixins to `range.front` to the program to
print the arrays:
[0, 5]
[1, 6]
[2, 7]
Ie I want to get something like this:
void walk(R)(R range) {
while (!range.empty) {
// mixin(`[idx ~ mixin(range.front)[1 .. $]);`);
range.popFront;
++idx;
}
}
Can I do this?
-
Thank you for the function `walk` Mark Isaacson of presentation
DConf 2015:
http://dconf.org/2015/talks/isaacson.pdf
Re: drastic slowdown for copies
On 05/29/2015 06:55 AM, Momo wrote: Perhaps you can give me another detailed answer. I get a slowdown for all parts (ref, copy and move) if I use uninitialized floats. Floating point variables are initialized to .nan of their types (e.g. float.nan). Apparently, the CPU is slow when using those special values: http://stackoverflow.com/questions/3606054/how-slow-is-nan-arithmetic-in-the-intel-x64-fpu Ali
