On Sunday, 22 October 2023 at 13:16:11 UTC, Imperatorn wrote:
On Sunday, 22 October 2023 at 13:13:55 UTC, Imperatorn wrote:
I would just like to share some knowledge about profiling an
exe using Visual Studio, since many might already have that
installed.
[...]
On Sunday, 22 October 2023
On Sunday, 22 October 2023 at 13:13:55 UTC, Imperatorn wrote:
I would just like to share some knowledge about profiling an
exe using Visual Studio, since many might already have that
installed.
[...]
On Sunday, 22 October 2023 at 13:13:55 UTC, Imperatorn wrote:
Just a clarification.
"open
On Tuesday, 24 August 2021 at 10:33:07 UTC, JG wrote:
The reason for the crash boils down to the fact that this fails:
foreach(k; sort!"a > b"(funcs.keys)) assert(k in funcs);
funcs is of type ubyte[4][float]
Is this a compiler bug?
That assert will fail if there are NaN keys in the AA.
On Tuesday, 24 August 2021 at 09:45:31 UTC, JG wrote:
On Tuesday, 24 August 2021 at 09:42:29 UTC, JG wrote:
On Tuesday, 24 August 2021 at 09:36:06 UTC, JG wrote:
[...]
In case anyone is interested it seems that it was forked to
here:
https://github.com/joakim-brannstrom/profdump
Perhaps
On Tuesday, 24 August 2021 at 09:42:29 UTC, JG wrote:
On Tuesday, 24 August 2021 at 09:36:06 UTC, JG wrote:
On Wednesday, 10 February 2021 at 23:42:31 UTC, mw wrote:
[...]
I tried to do this, but I am not sure how to install profdump.
What I did is cloned the repository using git.
Tried to
On Tuesday, 24 August 2021 at 09:36:06 UTC, JG wrote:
On Wednesday, 10 February 2021 at 23:42:31 UTC, mw wrote:
[...]
I tried to do this, but I am not sure how to install profdump.
What I did is cloned the repository using git.
Tried to build it using dub but got an error as described here:
On Wednesday, 10 February 2021 at 23:42:31 UTC, mw wrote:
On Wednesday, 10 February 2021 at 11:52:51 UTC, JG wrote:
As a follow up question I would like to know what tool people
use to profile d programs?
I use this one:
https://code.dlang.org/packages/profdump
e.g.
```
dub build
On 2/9/21 12:45 AM, JG wrote:
I was trying to profile a d program. So I ran: dub build
--build=profile. I then ran the program and it produced trace.log and
trace.def. I then ran d-profile-viewer and got the following error:
On Wednesday, 10 February 2021 at 23:42:31 UTC, mw wrote:
On Wednesday, 10 February 2021 at 11:52:51 UTC, JG wrote:
As a follow up question I would like to know what tool people
use to profile d programs?
I use this one:
https://code.dlang.org/packages/profdump
e.g.
```
dub build
On Wednesday, 10 February 2021 at 11:52:51 UTC, JG wrote:
As a follow up question I would like to know what tool people
use to profile d programs?
I use this one:
https://code.dlang.org/packages/profdump
e.g.
```
dub build --build=debug --build=profile
# run your program to generate
On Wednesday, 10 February 2021 at 13:31:09 UTC, Guillaume Piolat
wrote:
On Wednesday, 10 February 2021 at 11:52:51 UTC, JG wrote:
[...]
Here is what I use for sampling profiler:
(On Windows)
Build with LDC, x86_64, with dub -b release-debug in order to
have debug info.
Run your program
On Wednesday, 10 February 2021 at 11:52:51 UTC, JG wrote:
Thanks for the suggestions. However, I would prefer not to
spend time trying to debug d-profile-viewer at the moment.
As a follow up question I would like to know what tool people
use to profile d programs?
Here is what I use for
On 2/10/21 2:52 PM, JG wrote:
On Tuesday, 9 February 2021 at 18:33:16 UTC, drug wrote:
On Tuesday, 9 February 2021 at 07:45:13 UTC, JG wrote:
I was trying to profile a d program. So I ran: dub build
--build=profile. I then ran the program and it produced trace.log and
trace.def. I then ran
On Tuesday, 9 February 2021 at 18:33:16 UTC, drug wrote:
On Tuesday, 9 February 2021 at 07:45:13 UTC, JG wrote:
I was trying to profile a d program. So I ran: dub build
--build=profile. I then ran the program and it produced
trace.log and trace.def. I then ran d-profile-viewer and got
the
On Tuesday, 9 February 2021 at 07:45:13 UTC, JG wrote:
I was trying to profile a d program. So I ran: dub build
--build=profile. I then ran the program and it produced
trace.log and trace.def. I then ran d-profile-viewer and got
the following error:
On Tuesday, 9 February 2021 at 16:39:25 UTC, Dukc wrote:
You may have or may not have done it wrong, but in any case
this is a bug. If you do something wrong, the program should
tell you what you did wrong, instead of telling you that
character '-' does not belong to middle of a long int.
Oh
On Tuesday, 9 February 2021 at 07:45:13 UTC, JG wrote:
Is d-profile-viewer no longer working? Or did I do something
wrong?
You may have or may not have done it wrong, but in any case this
is a bug. If you do something wrong, the program should tell you
what you did wrong, instead of telling
On Wednesday, 4 December 2019 at 22:51:45 UTC, Steven
Schveighoffer wrote:
I localized that the leak was actually being caused by
websockets. I want to write down my experience because I did some
weird stuff which seems to be working but I want to learn how it
actually make sense and works.
On 12/4/19 5:04 PM, kerdemdemir wrote:
GC.sizeof(cast(void*)object) will be super useful. I will use that.
I also tried GC: --DRT-gcopt=profile:1 already. It provides so little
information.
I need to find out which member AA or array of which object is causing
this memory problem of mine.I
On Wednesday, 4 December 2019 at 15:38:36 UTC, Steven
Schveighoffer wrote:
On 12/4/19 3:10 AM, Erdem wrote:
I am used to have cool tools like valgrid massif to visualize
the memory usage from C++ but in D it seems I am blind folded
looking for the problem.
Until now I tried:
--vgc option
On 12/4/19 3:10 AM, Erdem wrote:
I am used to have cool tools like valgrid massif to visualize the memory
usage from C++ but in D it seems I am blind folded looking for the problem.
Until now I tried:
--vgc option which show million things which makes it not useful
--build=profile-gc
which
On Monday, 29 October 2018 at 10:14:23 UTC, Dukc wrote:
When I run the program, where is the performance profile file
supposed to appear? I can find nothing new in the
program/project root directory.
The same thing happens when I try to get a coverage report: No
file appears.
On Thursday, 1 November 2018 at 13:59:39 UTC, Guillaume Piolat
wrote:
On Monday, 29 October 2018 at 10:14:23 UTC, Dukc wrote:
I'm trying to profile my program, built like:
dub build --build=profile
When I run the program, where is the performance profile file
supposed to appear? I can find
On Monday, 29 October 2018 at 10:14:23 UTC, Dukc wrote:
I'm trying to profile my program, built like:
dub build --build=profile
When I run the program, where is the performance profile file
supposed to appear? I can find nothing new in the
program/project root directory. This happens
On Thursday, 27 July 2017 at 15:16:35 UTC, Eugene Wissner wrote:
Are there profilers that work well with dmd? valgrind? OProfile?
Yes, any sampling profiler works fine, e.g. perf on linux, Intel
VTune/AMD CodeXL on Windows.
Those directly monitor CPU performance counters and have a
On 2017-07-28 11:30, Mario Kröplin wrote:
Our programs are intended to run "forever". 24/7 servers.
What's wrong with having a bool that determines if the loop should
continue running?
--
/Jacob Carlborg
On Friday, 28 July 2017 at 09:02:10 UTC, Temtaime wrote:
There's no "gracefully" way to kill a thread.
If your thread cannot join, then you're doing something wrong
Our programs are intended to run "forever". 24/7 servers.
On Friday, 28 July 2017 at 08:06:33 UTC, Eugene Wissner wrote:
On Friday, 28 July 2017 at 06:32:59 UTC, Jacob Carlborg wrote:
On 2017-07-27 16:30, Eugene Wissner wrote:
I have a multi-threaded application, whose threads normally
run forever. But I need to profile this program, so I compile
On Friday, 28 July 2017 at 06:32:59 UTC, Jacob Carlborg wrote:
On 2017-07-27 16:30, Eugene Wissner wrote:
I have a multi-threaded application, whose threads normally
run forever. But I need to profile this program, so I compile
the code with -profile, send a SIGTERM and call exit(0) from
my
On 2017-07-27 16:30, Eugene Wissner wrote:
I have a multi-threaded application, whose threads normally run forever.
But I need to profile this program, so I compile the code with -profile,
send a SIGTERM and call exit(0) from my signal handler to exit the
program. The problem is that I get the
On Thursday, 27 July 2017 at 14:44:31 UTC, Temtaime wrote:
Also there was an issue that profiling doesn't work with
multi-threaded apps and leads to a crash.
Don't know if it is fixed.
Was fixed two years ago:
http://forum.dlang.org/post/mia2kf$djb$1...@digitalmars.com
On Thursday, 27 July 2017 at 14:52:18 UTC, Stefan Koch wrote:
On Thursday, 27 July 2017 at 14:30:33 UTC, Eugene Wissner wrote:
I have a multi-threaded application, whose threads normally
run forever. But I need to profile this program, so I compile
the code with -profile, send a SIGTERM and
On Thursday, 27 July 2017 at 14:30:33 UTC, Eugene Wissner wrote:
I have a multi-threaded application, whose threads normally run
forever. But I need to profile this program, so I compile the
code with -profile, send a SIGTERM and call exit(0) from my
signal handler to exit the program. The
Also there was an issue that profiling doesn't work with
multi-threaded apps and leads to a crash.
Don't know if it is fixed.
Exit is not "normal exit" for D programs so, do not use it.
Your threads should stop at some point to make able the app exit
successfully.
There's a "join" method. You can use it with your "done" variable.
On Monday, 17 July 2017 at 20:36:58 UTC, Igor wrote:
I also tried "Very Sleepy" profiler but it only shows symbols
for main application and not for the DLL that it loads which is
also built with debug info.
Something that works very well is CPU profiling with CodeXL. It
used to be an
On 17.07.2017 22:36, Igor wrote:
Is there a known limitation in profiling these or am I doing something
wrong?
When I try to run my application from VisualD (x64 build) with -profile
switch I just get Access Violation reported on WinMain function (actual
declaration, it doesn't enter its
I'm not sure if it's what happening in this case but, in code
as simple as this, function calls can sometimes be the
bottleneck. You should see how compiling with/without -O
affects performance, and adding `pragma(inline)` to funcB.
I guess my question is whether it is possible to have
I'm not sure if it's what happening in this case but, in code
as simple as this, function calls can sometimes be the
bottleneck. You should see how compiling with/without -O
affects performance, and adding `pragma(inline)` to funcB.
When compiled with -inline, the profiler does not report the
On Saturday, 21 January 2017 at 12:33:57 UTC, albert-j wrote:
Now I dmd -profile it and look at the performance of funcA with
d-profile-viewer. Inside funcA, only 20% of time is spend in
funcB, but the rest 80% is self-time of funcA. How is it
possible, when funcB has three times the
Awesome! These are exactly what I was looking for. Thanks!
On Saturday, 25 January 2014 at 15:34:57 UTC, Philippe Sigaud
wrote:
On Fri, Jan 24, 2014 at 10:25 PM, Benjamin Thaut
c...@benjamin-thaut.de wrote:
What Plattform are you profiling on?
Linux 32bits. Does it change something? I'm not using any
OS-specific
part of Phobos, AFAICT.
Doesn't
On Fri, Jan 24, 2014 at 10:25 PM, Benjamin Thaut c...@benjamin-thaut.de wrote:
What Plattform are you profiling on?
Linux 32bits. Does it change something? I'm not using any OS-specific
part of Phobos, AFAICT.
24-Jan-2014 20:31, Philippe Sigaud пишет:
I'm trying to use the `-profile` flag for DMD and, without any
documentation, I can't really understand the resulting log files:
* They contain only mangled names. Is there a way to get demangled,
human-readable symbols?
pipe it though ddemangle tool.
Please file a bug on this. The docs should be on dlang.org.
On 1/24/14 11:15 AM, Dmitry Olshansky wrote:
24-Jan-2014 20:31, Philippe Sigaud пишет:
I'm trying to use the `-profile` flag for DMD and, without any
documentation, I can't really understand the resulting log files:
* They contain
On Fri, Jan 24, 2014 at 8:35 PM, Brad Roberts bra...@puremagic.com wrote:
Please file a bug on this. The docs should be on dlang.org.
Here it is:
https://d.puremagic.com/issues/show_bug.cgi?id=11985
Am 24.01.2014 17:31, schrieb Philippe Sigaud:
I'm trying to use the `-profile` flag for DMD and, without any
documentation, I can't really understand the resulting log files:
* They contain only mangled names. Is there a way to get demangled,
human-readable symbols?
* Can someone tell me what
On 07/31/2013 02:31 PM, bearophile wrote:
If this resulting array is used only for a short period of time:
return sequence.map!(x = x.func).array;
Just to compare, I tried rewriting the cacheing version of the neighbours()
function to use this kind of sequence. Here's the original code:
On 07/31/2013 01:02 AM, bearophile wrote:
Also try to use a bitvector from Phobos instead of bool[].
I take it you mean std.bitmanip.BitArray ... ?
I ran into some interesting problems which are probably down to incomplete
implementation: e.g.
myBitArray.length += n;
... results in a
Joseph Rushton Wakeling:
I take it you mean std.bitmanip.BitArray ... ?
Right.
I ran into some interesting problems which are probably down to
incomplete
implementation: e.g.
myBitArray.length += n;
... results in a compiler error message: myBitArray.length() is
not an lvalue
On 07/31/2013 02:31 PM, bearophile wrote:
Then I suggest to try ldc2 too.
I haven't profiled with ldc2 yet (I will) but the broad speed tendencies are
similar. It's faster but not so much faster that I can assume the speed hit
isn't still there.
My suggestion is to copy the implementation of
Joseph Rushton Wakeling:
But most likely the best compromise solution is as you
suggest a buffer, with the instruction .dup the result
if you want it to be persistent.
In my library code often I do the opposite: I add a template
boolean switch named as doCopy that defaults to true. If it's
There are many ways to design a graph data structure. All of them
have tradeoffs, they can't be very good for everything.
For your benchmarks are you using the ldc2 compiler with correct
compilation switches?
map() should not allocate memory.
Sometimes lazy code is faster and sometimes it's
On Tuesday, 30 July 2013 at 23:02:40 UTC, bearophile wrote:
Sometimes you can keep memory allocation low, performance
almost acceptable, and code easy to read using a pre-allocated
scratch space plus using map() and copy() from std.algorithm.
Any chance of some nice examples of this?
John Colvin:
Any chance of some nice examples of this?
Allocate a buffer somewhere, on the stack or heap, and then
instead of using a map()+array() as in that code use map()+copy()
plus a test to be sure the space is sufficient.
But in the end I don't know how much well even ldc2 will
On 07/31/2013 01:02 AM, bearophile wrote:
There are many ways to design a graph data structure. All of them have
tradeoffs, they can't be very good for everything.
Sure, it's just that in this case the reference code (igraph) is _much_ more
performant. I'm keen to try and identify the major
On Friday, December 07, 2012 19:49:30 js.mdnq wrote:
Is there a standard library that makes profiling easy and
accurate?
e.g.,
profile.start();
...
writeln(profile.end());
where profile.end() can has some nicely printed string to
output(like xxm xxs xxms)?
The compiler has a
Sorry, I just saw the generated file...
:p
58 matches
Mail list logo
