Re: GDB + ddemangle
On Friday, 26 April 2019 at 07:08:45 UTC, Arun Chandrasekaran wrote: On Friday, 20 April 2018 at 17:55:12 UTC, Iain Buclaw wrote: On 20 April 2018 at 17:40, drug via Digitalmars-d-announce wrote: 20.04.2018 16:49, Iain Buclaw пишет: [...] it works, thank you. But not in all cases. For example when gdb stops on breakpoint it demangle, but if I do `bt` - backtrace isn't demangled. Using a compiler that implements 2.077 or later (IIRC) probably won't, due to gdb being too old. They broke ABI by introducing back referencing, no release of gdb supports that yet. This stays broken after 1 year. How do we fix this? I recognize your name from IRC, where you posted a related question yesterday. The person in charge for D in GDB had answered: The problem is that the mangling scheme has changed (to produce shorter mangles). Demangling will be eventually supported when D will be in GCC for good (GCC 10) and then he would port the demangling from GCC to GDB. So for now demangling of D in GDB only work if the binary is compiled with a very old version of the compiler, i.e prior to the new mangling method.
Re: Beta 2.086.0
On Saturday, 20 April 2019 at 14:16:09 UTC, Martin Nowak wrote: [...] __traits for private symbols AND copy constructors?!?! Awesome!
Re: DMD metaprogramming enhancement
On Thursday, 25 April 2019 at 23:41:32 UTC, Suleyman wrote:
Hello everyone,
I am happy to announce that in the next DMD release you will be
able to more freely enjoy your metaprograming experience now
that a long-standing limitation has been lifted.
You can now instantiate local and member templates with local
symbols.
Example:
---
struct S
{
private int _m;
void exec(alias fun)()
{
fun(_m);
}
}
unittest
{
int localVar;
void set(int i)
{
localVar = i;
}
auto obj = S(10);
obj.exec!set(); // no error or warning
assert(localVar == 10);
}
---
I hope you enjoy!
Thanks a lot. It is really refreshing to know this has been fixed.
Re: grain - D Language for Deep Learning
On Friday, 26 April 2019 at 06:35:42 UTC, Shigeki Karita wrote: I haven't know that GPU support in Stan. That's Cool! Cholesky decomposition always suffers me when I use covariance matrix or something. If you are interested in GPU acceleration in probabilistic programming, see also this paper (Table 2) of Edward (previous name of Tensorflow Probability) https://arxiv.org/pdf/1701.03757.pdf I think I recall hearing something about Edward. In my experience, Bayesian modelling can be quite finicky...you might do something to get faster results, but then the results may not make sense, particularly as the model becomes more complicated. While I often prefer the Bayesian approach, faster doesn't necessarily mean better.
Re: DStep 1.0.0
On 2019-04-22 11:02:24 +, Jacob Carlborg said: ... and support for one more platform has been added: Windows... Are there are any functional differences between the platforms? Or can I just use the OSX version and use the generated .d files with the DMD Windows version too? -- Robert M. Münch http://www.saphirion.com smarter | better | faster
Re: GDB + ddemangle
On Friday, 20 April 2018 at 17:55:12 UTC, Iain Buclaw wrote: On 20 April 2018 at 17:40, drug via Digitalmars-d-announce wrote: 20.04.2018 16:49, Iain Buclaw пишет: [...] it works, thank you. But not in all cases. For example when gdb stops on breakpoint it demangle, but if I do `bt` - backtrace isn't demangled. Using a compiler that implements 2.077 or later (IIRC) probably won't, due to gdb being too old. They broke ABI by introducing back referencing, no release of gdb supports that yet. This stays broken after 1 year. How do we fix this?
Re: grain - D Language for Deep Learning
On Wednesday, 24 April 2019 at 17:31:03 UTC, jmh530 wrote: On Wednesday, 24 April 2019 at 16:33:00 UTC, Shigeki Karita wrote: [snip] I see. I'm interested in Stan that is the best library for probabilistic models but it lacks of GPU computation. Therefore, I plan to add some probabilistic programming paradigm into grain like pytorch (pyro) and tensorflow (tf probability). Conveniently enough, they just incorporated some GPU support in the release in March [1]. Here's an earlier status update [2]. The initial work was focused on cholesky decomposition because that was a big source of slowdown for some types of models. Probably still has a ways to go before reaching tensorflows maturity on the GPU. [1] https://github.com/stan-dev/math/releases/tag/v2.19.0 [2] https://discourse.mc-stan.org/t/gpu-update-whats-up-and-where-we-are-going/6015 I haven't know that GPU support in Stan. That's Cool! Cholesky decomposition always suffers me when I use covariance matrix or something. If you are interested in GPU acceleration in probabilistic programming, see also this paper (Table 2) of Edward (previous name of Tensorflow Probability) https://arxiv.org/pdf/1701.03757.pdf
Re: DMD metaprogramming enhancement
On Friday, 26 April 2019 at 06:29:04 UTC, Simen Kjærås wrote:
On Thursday, 25 April 2019 at 23:41:32 UTC, Suleyman wrote:
Hello everyone,
I am happy to announce that in the next DMD release you will
be able to more freely enjoy your metaprograming experience
now that a long-standing limitation has been lifted.
You can now instantiate local and member templates with local
symbols.
Example:
---
struct S
{
private int _m;
void exec(alias fun)()
{
fun(_m);
}
}
unittest
{
int localVar;
void set(int i)
{
localVar = i;
}
auto obj = S(10);
obj.exec!set(); // no error or warning
assert(localVar == 10);
}
---
I hope you enjoy!
You have no idea how happy I am to hear this has been fixed! So
many of my designs have been hamstrung by 5710, and it's been
around since the dawn of time.
BTW, at least two people have promised money outside BountySource
to have 5710 fixed:
https://forum.dlang.org/post/gjzrklkxfmgjjdfor...@forum.dlang.org
--
Simen
Re: DMD metaprogramming enhancement
On Thursday, 25 April 2019 at 23:41:32 UTC, Suleyman wrote:
Hello everyone,
I am happy to announce that in the next DMD release you will be
able to more freely enjoy your metaprograming experience now
that a long-standing limitation has been lifted.
You can now instantiate local and member templates with local
symbols.
Example:
---
struct S
{
private int _m;
void exec(alias fun)()
{
fun(_m);
}
}
unittest
{
int localVar;
void set(int i)
{
localVar = i;
}
auto obj = S(10);
obj.exec!set(); // no error or warning
assert(localVar == 10);
}
---
I hope you enjoy!
You have no idea how happy I am to hear this has been fixed! So
many of my designs have been hamstrung by 5710, and it's been
around since the dawn of time.
--
Simen
Re: DMD metaprogramming enhancement
On Thursday, 25 April 2019 at 23:41:32 UTC, Suleyman wrote:
Hello everyone,
I am happy to announce that in the next DMD release you will be
able to more freely enjoy your metaprograming experience now
that a long-standing limitation has been lifted.
You can now instantiate local and member templates with local
symbols.
Example:
---
struct S
{
private int _m;
void exec(alias fun)()
{
fun(_m);
}
}
unittest
{
int localVar;
void set(int i)
{
localVar = i;
}
auto obj = S(10);
obj.exec!set(); // no error or warning
assert(localVar == 10);
}
---
I hope you enjoy!
Noice! Finally indeed. thank you! :D
